Download complete issue (pdf 3360kb) - Academic Journals
Download complete issue (pdf 3360kb) - Academic Journals
Download complete issue (pdf 3360kb) - Academic Journals
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Scientific Research<br />
and Essays<br />
Volume 7 Number 18 16 May, 2012<br />
ISSN 1992-2248
ABOUT SRE<br />
The Scientific Research and Essays (SRE) is published weekly (one volume per year) by <strong>Academic</strong> <strong>Journals</strong>.<br />
Scientific Research and Essays (SRE) is an open access journal with the objective of publishing quality research<br />
articles in science, medicine, agriculture and engineering such as Nanotechnology, Climate Change and Global<br />
Warming, Air Pollution Management and Electronics etc. All papers published by SRE are blind peer reviewed.<br />
Submission of Manuscript<br />
Submit manuscripts as e-mail attachment to the Editorial Office at: sre@cademicjournals.org,<br />
sre.research.journal@gmail.com. A manuscript number will be mailed to the corresponding author shortly<br />
after submission.<br />
For all other correspondence that cannot be sent by e-mail, please contact the editorial office for the<br />
appropriate address.<br />
The Scientific Research and Essays will only accept manuscripts submitted as e-mail attachments.<br />
Please read the Instructions for Authors before submitting your manuscript. The manuscript files should be<br />
given the last name of the first author.
Editors<br />
Dr. NJ Tonukari<br />
Editor-in-Chief<br />
Scientific Research and Essays<br />
<strong>Academic</strong> <strong>Journals</strong><br />
E-mail: sre.research.journal@gmail.com<br />
Dr. M. Sivakumar Ph.D. (Tech).<br />
Associate Professor<br />
School of Chemical & Environmental Engineering<br />
Faculty of Engineering<br />
University of Nottingham<br />
Jalan Broga, 43500 Semenyih<br />
Selangor Darul Ehsan<br />
Malaysia.<br />
Prof. N. Mohamed El Sawi Mahmoud<br />
Department of Biochemistry, Faculty of science,<br />
King AbdulAziz university,<br />
Saudia Arabia.<br />
Prof. Ali Delice<br />
Science and Mathematics Education Department,<br />
Atatürk Faculty of Education,<br />
Marmara University,<br />
Turkey.<br />
Prof. Mira Grdisa<br />
Rudjer Boskovic Institute, Bijenicka cesta 54,<br />
Croatia.<br />
Prof. Emmanuel Hala Kwon-Ndung<br />
Nasarawa State University Keffi Nigeria<br />
PMB 1022 Keffi,<br />
Nasarawa State.<br />
Nigeria.<br />
Dr. Cyrus Azimi<br />
Department of Genetics, Cancer Research Center,<br />
Cancer Institute, Tehran University of Medical Sciences,<br />
Keshavarz Blvd.,<br />
Tehran, Iran.<br />
Dr. Gomez, Nidia Noemi<br />
National University of San Luis,<br />
Faculty of Chemistry, Biochemistry and Pharmacy,<br />
Laboratory of Molecular Biochemistry Ejercito de los<br />
Andes 950 - 5700 San Luis<br />
Argentina.<br />
Prof. M. Nageeb Rashed<br />
Chemistry Department- Faculty of Science, Aswan<br />
South Valley University,<br />
Egypt.<br />
Dr. John W. Gichuki<br />
Kenya Marine & Fisheries Research Institute,<br />
Kenya.<br />
Dr. Wong Leong Sing<br />
Department of Civil Engineering,<br />
College of Engineering,<br />
Universiti Tenaga Nasional,<br />
Km 7, Jalan Kajang-Puchong,<br />
43009 Kajang, Selangor Darul Ehsan,<br />
Malaysia.<br />
Prof. Xianyi LI<br />
College of Mathematics and Computational Science<br />
Shenzhen University<br />
Guangdong, 518060<br />
P.R. China.<br />
Prof. Mevlut Dogan<br />
Kocatepe University, Science Faculty,<br />
Physics Dept. Afyon/ Turkey.<br />
Turkey .<br />
Prof. Kwai-Lin Thong<br />
Microbiology Division,<br />
Institute of Biological Science,<br />
Faculty of Science,University of Malaya,<br />
50603, Kuala Lumpur,<br />
Malaysia.<br />
Prof. Xiaocong He<br />
Faculty of Mechanical and Electrical Engineering,<br />
Kunming University of Science and Technology,<br />
253 Xue Fu Road, Kunming,<br />
P.R. China.<br />
Prof. Sanjay Misra<br />
Department of Computer Engineering<br />
School of Information and Communication Technology<br />
Federal University of Technology, Minna,<br />
Nigeria.<br />
Prof. Burtram C. Fielding Pr.Sci.Nat.<br />
Department of Medical BioSciences<br />
University of the Western Cape<br />
Private Bag X17<br />
Modderdam Road<br />
Bellville, 7535,<br />
South Africa.<br />
Prof. Naqib Ullah Khan<br />
Department of Plant Breeding and Genetics<br />
NWFP Agricultural University Peshawar 25130,<br />
Pakistan
Editorial Board<br />
Prof. Ahmed M. Soliman<br />
20 Mansour Mohamed St.,Apt 51,<br />
Zamalek, Cairo,<br />
Egypt.<br />
Prof. Juan José Kasper Zubillaga<br />
Av. Universidad 1953 Ed. 13 depto 304,<br />
México D.F. 04340,<br />
México.<br />
Prof. Chau Kwok-wing<br />
University of Queensland<br />
Instituto Mexicano del Petroleo,<br />
Eje Central Lazaro Cardenas<br />
Mexico D.F.,<br />
Mexico.<br />
Prof. Raj Senani<br />
Netaji Subhas Institute of Technology,<br />
Azad Hind Fauj Marg,<br />
Sector 3,<br />
Dwarka, New Delhi 110075,<br />
India.<br />
Prof. Robin J Law<br />
Cefas Burnham Laboratory,<br />
Remembrance AvenueBurnham on Crouch,<br />
Essex CM0 8HA,<br />
UK.<br />
Prof. V. Sundarapandian<br />
Indian Institute of Information Technology and<br />
Management-Kerala<br />
Park Centre,<br />
Technopark Campus, Kariavattom P.O.,<br />
Thiruvananthapuram-695 581, Kerala,<br />
India.<br />
Prof. Tzung-Pei Hong<br />
Department of Electrical Engineering,<br />
and at the Department of Computer Science and<br />
Information Engineering<br />
National University of Kaohsiung.<br />
Prof. Zulfiqar Ahmed<br />
Department of Earth Sciences, box 5070,<br />
Kfupm,dhahran - 31261,<br />
Saudi Arabia.<br />
Prof. Khalifa Saif Al-Jabri<br />
Department of Civil and Architectural Engineering<br />
College of Engineering,<br />
Sultan Qaboos University<br />
P.O. Box 33, Al-Khod 123, Muscat.<br />
Sultanate of Oman.<br />
Prof. Konstantinos<br />
43 Liakateon St.11474 Athens,<br />
Greece<br />
Prof. Liotta Leonarda<br />
Prof. V. Sundarapandian<br />
Indian Institute of Information Technology & Management -<br />
Kerala<br />
Park Centre,<br />
Technopark, Kariavattom P.O.<br />
Thiruvananthapuram-695 581,<br />
Kerala India.<br />
Prof. Thangavelu Perianan<br />
Department of Mathematics, Aditanar College,<br />
Tiruchendur-628216 India.<br />
Prof. Yan-ze Peng<br />
Department of Mathematics,<br />
Huazhong University of Science and Technology,<br />
Wuhan 430074, P. R.<br />
China.<br />
Prof. Konstantinos D. Karamanos<br />
Universite Libre de Bruxelles,<br />
CP 231 Centre of Nonlinear Phenomena<br />
And Complex systems,<br />
CENOLI Boulevard de Triomphe<br />
B-1050,<br />
Brussels, Belgium.<br />
Prof. Xianyi LI<br />
School of Mathematics and Physics,<br />
Nanhua University, Hengyang City,<br />
Hunan Province,<br />
P. R. China.<br />
Dr. K.W. Chau<br />
Hong Kong Polytechnic University<br />
Department of Civil & Structural Engineering,<br />
Hong Kong Polytechnic University, Hunghom,<br />
Kowloon, Hong Kong,<br />
China.<br />
Dr. Amadou Gaye<br />
LPAO-SF / ESP Po Box 5085 Dakar-Fann SENEGAL<br />
University Cheikh Anta Diop Dakar<br />
SENEGAL.<br />
Prof. Masno Ginting<br />
P2F-LIPI, Puspiptek-Serpong,<br />
15310 Indonesian Institute of Sciences,<br />
Banten-Indonesia.<br />
Dr. Ezekiel Olukayode Idowu<br />
Department of Agricultural Economics,<br />
Obafemi Awolowo University,<br />
Ife-Ife,<br />
Nigeria.<br />
Dr. Fei Kang<br />
Faculty of Infrastructure Engineering,<br />
Dalian University of Technology,<br />
Dalian, 116024, China<br />
Dr. Jane Wanjiku Kabubo-Mariara
Electronic submission of manuscripts is strongly<br />
encouraged, provided that the text, tables, and figures are<br />
included in a single Microsoft Word file (preferably in Arial<br />
font).<br />
The cover letter should include the corresponding author's<br />
full address and telephone/fax numbers and should be in<br />
an e-mail message sent to the Editor, with the file, whose<br />
name should begin with the first author's surname, as an<br />
attachment.<br />
Article Types<br />
Three types of manuscripts may be submitted:<br />
Regular articles: These should describe new and carefully<br />
confirmed findings, and experimental procedures should<br />
be given in sufficient detail for others to verify the work.<br />
The length of a full paper should be the minimum required<br />
to describe and interpret the work clearly.<br />
Short Communications: A Short Communication is suitable<br />
for recording the results of <strong>complete</strong> small investigations<br />
or giving details of new models or hypotheses, innovative<br />
methods, techniques or apparatus. The style of main<br />
sections need not conform to that of full-length papers.<br />
Short communications are 2 to 4 printed pages (about 6 to<br />
12 manuscript pages) in length.<br />
Reviews: Submissions of reviews and perspectives covering<br />
topics of current interest are welcome and encouraged.<br />
Reviews should be concise and no longer than 4-6 printed<br />
pages (about 12 to 18 manuscript pages). Reviews are also<br />
peer-reviewed.<br />
Review Process<br />
Instructions for Author<br />
All manuscripts are reviewed by an editor and members of<br />
the Editorial Board or qualified outside reviewers. Authors<br />
cannot nominate reviewers. Only reviewers randomly<br />
selected from our database with specialization in the<br />
subject area will be contacted to evaluate the manuscripts.<br />
The process will be blind review.<br />
Decisions will be made as rapidly as possible, and the<br />
journal strives to return reviewers’ comments to authors as<br />
fast as possible. The editorial board will re-review<br />
manuscripts that are accepted pending revision. It is the<br />
goal of the SRE to publish manuscripts within weeks after<br />
submission.<br />
Regular articles<br />
All portions of the manuscript must be typed doublespaced<br />
and all pages numbered starting from the title<br />
page.<br />
The Title should be a brief phrase describing the<br />
contents of the paper. The Title Page should include the<br />
authors' full names and affiliations, the name of the<br />
corresponding author along with phone, fax and E-mail<br />
information. Present addresses of authors should<br />
appear as a footnote.<br />
The Abstract should be informative and <strong>complete</strong>ly selfexplanatory,<br />
briefly present the topic, state the scope of<br />
the experiments, indicate significant data, and point out<br />
major findings and conclusions. The Abstract should be<br />
100 to 200 words in length.. Complete sentences, active<br />
verbs, and the third person should be used, and the<br />
abstract should be written in the past tense. Standard<br />
nomenclature should be used and abbreviations should<br />
be avoided. No literature should be cited.<br />
Following the abstract, about 3 to 10 key words that will<br />
provide indexing references should be listed.<br />
A list of non-standard Abbreviations should be added.<br />
In general, non-standard abbreviations should be used<br />
only when the full term is very long and used often.<br />
Each abbreviation should be spelled out and introduced<br />
in parentheses the first time it is used in the text. Only<br />
recommended SI units should be used. Authors should<br />
use the solidus presentation (mg/ml). Standard<br />
abbreviations (such as ATP and DNA) need not be<br />
defined.<br />
The Introduction should provide a clear statement of<br />
the problem, the relevant literature on the subject, and<br />
the proposed approach or solution. It should be<br />
understandable to colleagues from a broad range of<br />
scientific disciplines.<br />
Materials and methods should be <strong>complete</strong> enough<br />
to allow experiments to be reproduced. However, only<br />
truly new procedures should be described in detail;<br />
previously published procedures should be cited, and<br />
important modifications of published procedures should<br />
be mentioned briefly. Capitalize trade names and<br />
include the manufacturer's name and address.<br />
Subheadings should be used. Methods in general use<br />
need not be described in detail.
Results should be presented with clarity and precision.<br />
The results should be written in the past tense when<br />
describing findings in the authors' experiments.<br />
Previously published findings should be written in the<br />
present tense. Results should be explained, but largely<br />
without referring to the literature. Discussion,<br />
speculation and detailed interpretation of data should<br />
not be included in the Results but should be put into the<br />
Discussion section.<br />
The Discussion should interpret the findings in view of<br />
the results obtained in this and in past studies on this<br />
topic. State the conclusions in a few sentences at the end<br />
of the paper. The Results and Discussion sections can<br />
include subheadings, and when appropriate, both<br />
sections can be combined.<br />
The Acknowledgments of people, grants, funds, etc<br />
should be brief.<br />
Tables should be kept to a minimum and be designed to<br />
be as simple as possible. Tables are to be typed doublespaced<br />
throughout, including headings and footnotes.<br />
Each table should be on a separate page, numbered<br />
consecutively in Arabic numerals and supplied with a<br />
heading and a legend. Tables should be self-explanatory<br />
without reference to the text. The details of the methods<br />
used in the experiments should preferably be described<br />
in the legend instead of in the text. The same data should<br />
not be presented in both table and graph form or<br />
repeated in the text.<br />
Figure legends should be typed in numerical order on a<br />
separate sheet. Graphics should be prepared using<br />
applications capable of generating high resolution GIF,<br />
TIFF, JPEG or Powerpoint before pasting in the Microsoft<br />
Word manuscript file. Tables should be prepared in<br />
Microsoft Word. Use Arabic numerals to designate<br />
figures and upper case letters for their parts (Figure 1).<br />
Begin each legend with a title and include sufficient<br />
description so that the figure is understandable without<br />
reading the text of the manuscript. Information given in<br />
legends should not be repeated in the text.<br />
References: In the text, a reference identified by means<br />
of an author‘s name should be followed by the date of<br />
the reference in parentheses. When there are more than<br />
two authors, only the first author‘s name should be<br />
mentioned, followed by ’et al‘. In the event that an<br />
author cited has had two or more works published during<br />
the same year, the reference, both in the text and in the<br />
reference list, should be identified by a lower case letter<br />
like ’a‘ and ’b‘ after the date to distinguish the works.<br />
Examples:<br />
Abayomi (2000), Agindotan et al. (2003), (Kelebeni,<br />
1983), (Usman and Smith, 1992), (Chege, 1998;<br />
Chukwura, 1987a,b; Tijani, 1993, 1995), (Kumasi et al.,<br />
2001)<br />
References should be listed at the end of the paper in<br />
alphabetical order. Articles in preparation or articles<br />
submitted for publication, unpublished observations,<br />
personal communications, etc. should not be included<br />
in the reference list but should only be mentioned in<br />
the article text (e.g., A. Kingori, University of Nairobi,<br />
Kenya, personal communication). Journal names are<br />
abbreviated according to Chemical Abstracts. Authors<br />
are fully responsible for the accuracy of the references.<br />
Examples:<br />
Ogunseitan OA (1998). Protein method for<br />
investigating mercuric reductase gene expression in<br />
aquatic environments. Appl. Environ. Microbiol.<br />
64:695–702.<br />
Gueye M, Ndoye I, Dianda M, Danso SKA, Dreyfus B<br />
(1997). Active N2 fixation in several Faidherbia albida<br />
provenances. Ar. Soil Res. Rehabil. 11:63-70.<br />
Charnley AK (1992). Mechanisms of fungal<br />
pathogenesis in insects with particular reference to<br />
locusts. In: Lomer CJ, Prior C (eds) Biological Controls of<br />
Locusts and Grasshoppers: Proceedings of an<br />
international workshop held at Cotonou, Benin. Oxford:<br />
CAB International, pp 181-190.<br />
Mundree SG, Farrant JM (2000). Some physiological<br />
and molecular insights into the mechanisms of<br />
desiccation tolerance in the resurrection plant<br />
Xerophyta viscasa Baker. In Cherry et al. (eds) Plant<br />
tolerance to abiotic stresses in Agriculture: Role of<br />
Genetic Engineering, Kluwer <strong>Academic</strong> Publishers,<br />
Netherlands, pp 201-222.<br />
Babalola OO (2002). Interactions between Striga<br />
hermonthica (Del.) Benth. and fluorescent rhizosphere<br />
bacteria Of Zea mays, L. and Sorghum bicolor L.<br />
Moench for Striga suicidal germination In Vigna<br />
unguiculata . PhD dissertation, University of Ibadan,<br />
Ibadan, Nigeria.<br />
Short Communications<br />
A Short Communication is suitable for recording the<br />
results of <strong>complete</strong> small investigations or giving<br />
details of new models or hypotheses, innovative<br />
methods, techniques or apparatus. The style of main<br />
sections need not conform to that of full-length papers.<br />
Short communications are 2 to 4 printed pages (about<br />
6 to 12 manuscript pages) in length.
Fees and Charges: Authors are required to pay a $550 handling fee. Publication of an article in the Scientific Research<br />
and Essays is not contingent upon the author's ability to pay the charges. Neither is acceptance to pay the handling<br />
fee a guarantee that the paper will be accepted for publication. Authors may still request (in advance) that the<br />
editorial office waive some of the handling fee under special circumstances.<br />
Copyright: © 2012, <strong>Academic</strong> <strong>Journals</strong>.<br />
All rights Reserved. In accessing this journal, you agree that you will access the contents for your own personal use<br />
but not for any commercial use. Any use and or copies of this Journal in whole or in part must include the customary<br />
bibliographic citation, including author attribution, date and article title.<br />
Submission of a manuscript implies: that the work described has not been published before (except in the form of an<br />
abstract or as part of a published lecture, or thesis) that it is not under consideration for publication elsewhere; that if<br />
and when the manuscript is accepted for publication, the authors agree to automatic transfer of the copyright to the<br />
publisher.<br />
Disclaimer of Warranties<br />
In no event shall <strong>Academic</strong> <strong>Journals</strong> be liable for any special, incidental, indirect, or consequential damages of any<br />
kind arising out of or in connection with the use of the articles or other material derived from the SRE, whether or not<br />
advised of the possibility of damage, and on any theory of liability.<br />
This publication is provided "as is" without warranty of any kind, either expressed or implied, including, but not<br />
limited to, the implied warranties of merchantability, fitness for a particular purpose, or non-infringement.<br />
Descriptions of, or references to, products or publications does not imply endorsement of that product or publication.<br />
While every effort is made by <strong>Academic</strong> <strong>Journals</strong> to see that no inaccurate or misleading data, opinion or statements<br />
appear in this publication, they wish to make it clear that the data and opinions appearing in the articles and<br />
advertisements herein are the responsibility of the contributor or advertiser concerned. <strong>Academic</strong> <strong>Journals</strong> makes no<br />
warranty of any kind, either express or implied, regarding the quality, accuracy, availability, or validity of the data or<br />
information in this publication or of any other publication to which it may be linked.
International Journal<br />
Scientific<br />
of Medicine<br />
Research<br />
and<br />
and<br />
Medical<br />
Essays<br />
Sciences<br />
Table of Contents: Volume 7 Number 18 16 May, 2012<br />
es<br />
ARTICLES<br />
Gastroprotective effects of Dicranopteris linearis leaf extract against<br />
ethanol-induced gastric mucosal injury in rats 1761<br />
Jamal Hussaini, Nurul Asyikin Othman, Mahmood Ameen Abdulla, Nazia<br />
Abdul Majid, Halabi Mohd Faroq and Salmah Ismail<br />
Introducing slag powder as drag reduction agent in pipeline: An<br />
experimental approach 1768<br />
Hayder A. Abdulbari, Siti Nuraffini Bt Kamarulizam, Rosli M. Y. and<br />
Arun Gupta<br />
Available transfer capability and least square method 1777<br />
Mojgan Hojabri, Mohammadsoroush Soheilirad and Mahdi Hedayati<br />
Seasonality of non-timber forest products in the Kupe mountain region<br />
of South Cameroon 1786<br />
Ngane B. K., Ngane E. B., Sumbele S. A., Njukeng J. N., Ngone M. A. and<br />
Ehabe E. E.<br />
Adaptive message size routing strategy for delay tolerant network 1798<br />
Qaisar Ayub, M. Soperi Mohd Zahid, Sulma Rashid and Abdul Hanan Abdullah<br />
Length-weight relationships, relative condition factor and relative weight<br />
of three fish species from beach seine fishing grounds in Iranian coastal<br />
waters of Caspian Sea 1809<br />
Moradinasab, GH., Raeisi, H, Paighambari, S. Y., Ghorbani, R and Bibak, Z.<br />
Design and development of standalone DSP prototype for QT interval<br />
processing and monitoring 1813<br />
Goh Chun Seng, Sh-Hussain Salleh, J. M. Najeb, I. Kamarulafizam,<br />
Mahyar Hamedi and Alias Md Noor<br />
Effective techniques in drilling to improve the recovery process of<br />
hydrocarbons in oil and gas sector 1830<br />
Shafqat Hameed and Muhammad Abbas Choudharyand Alias Md Noor
Scientific Research and Essays Vol. 7(18), pp. 1761-1767, 16 May, 2012<br />
Available online at http://www.academicjournals.org/SRE<br />
DOI: 10.5897/SRE11.775<br />
ISSN 1992-2248 ©2012 <strong>Academic</strong> <strong>Journals</strong><br />
Full Length Research Paper<br />
Gastroprotective effects of Dicranopteris linearis leaf<br />
extract against ethanol-induced gastric mucosal<br />
injury in rats<br />
Jamal Hussaini 1 *, Nurul Asyikin Othman 1 , Mahmood Ameen Abdulla 2 , Nazia Abdul Majid 3 ,<br />
Halabi Mohd Faroq 3 and Salmah Ismail 2<br />
1 Faculty of Medicine, Universiti Teknologi MARA (UiTM), 40450 Shah Alam, Selangor, Malaysia.<br />
2 Department of Molecular Medicine, Faculty of Medicine, University of Malaya, 50603, Malaysia.<br />
3 Institute of Biological Science, Faculty of Science, University of Malaya, 50603, Malaysia.<br />
Accepted 9 May, 2012<br />
Dicranopteris linearis is a medicinal plant commonly used traditionally in the treatment of many<br />
ailments. This study was performed to evaluate the gastroprotective effect of ethanolic extracts of D.<br />
linearis leaf extract (DLELE) against ethanol-induced gastric mucosal injury in experimental rats. The<br />
rats were divided into four groups respectively pre-treated orally with carboxymethyl cellulose (CMC)<br />
solution (ulcer control groups), omeprazole 20 mg/kg (reference group), 250 and 500 mg/kg of DLELE<br />
(experimental groups) one hour before oral administration of absolute ethanol to generate gastric<br />
mucosal damage. After an additional hour, the rats were sacrificed and the ulcer areas of the gastric<br />
walls were determined. The ulcer control group exhibited severe mucosal injury, whereas groups pretreated<br />
with DLELE exhibited significant protection of gastric mucosa. These findings were also<br />
confirmed by histology of gastric wall. Significant increases in gastric mucus production and decrease<br />
in acidity of gastric content were observed in treated groups with DLELE compare to ulcer control<br />
group. In conclusion, treatment with DLELE prior to absolute alcohol has significantly protect gastric<br />
mucosa as ascertained grossly by significant reduction of ulcer area, increases in gastric mucus<br />
production and decrease the acidity of gastric content and histology by comparatively decreases in<br />
gastric mucosal injury, reduction or absence of edema and leucocytes infiltration of submucosal layer<br />
compared to ulcer control group. DLELE was able to decrease the acidity and increase the mucosal<br />
defense in the gastric area, thereby justifying its use as an antiulcerogenic agent.<br />
Key words: Dicranopteris linearis, cytoprotection, gastric ulcer, mucus, histology.<br />
INTRODUCTION<br />
Gastric ulcer is an illness that affects a considerable<br />
number of people worldwide. The etiological factors of<br />
this disorder include stress, smoking, nutritional<br />
deficiencies, infections, frequent and indiscriminate use<br />
of nonsteroidal anti-inflammatory drugs (NSAIDs)<br />
(Khazaei and Salehi, 2006). The pathogenesis of<br />
*Corresponding author. E-mail: jamalum2004@yahoo.com,<br />
jamalh@salam.uitm.edu.my. Tel: +603-55442886. Fax: +603-<br />
55442831.<br />
gastroduodenal ulcers are influenced by various<br />
aggressive and defensive factors, such as mucus<br />
secretion, mucosal barrier, acid-pepsin secretion, blood<br />
flow, cellular regeneration and endogenous protective<br />
agents (prostaglandins and epidermal growth factor<br />
(Mizui et al., 1987). Although, the introduction of protonpump<br />
inhibitors to the classic anti-ulcer therapy had<br />
revolutionized treatment of peptic ulcers and other<br />
gastrointestinal disorders, there is still no <strong>complete</strong> cure<br />
for this disease. It has been shown that long term use of<br />
these drugs leads to various adverse and side effects.<br />
Relapses of the malady, ineffectiveness of different drug
1762 Sci. Res. Essays<br />
regimens and even resistance to drugs are emerging (Al-<br />
Mofleh et al., 2007). Thus, there is an urgent requirement<br />
to identify more effective and safe anti-ulcer agents. A<br />
widespread search has been launched to identify new<br />
anti-ulcer therapies from natural sources. Herbs,<br />
medicinal plants, spices, vegetables and crude drug<br />
substances are considered to be a potential source to<br />
combat various diseases including gastric ulcer. In the<br />
scientific literature, a large number of medicinal plants<br />
with gastric anti-ulcer potential have been reported<br />
(Abdulla et al., 2010; Ketuly et al., 2011; Mahmood et al.,<br />
2010; Wasman et al., 2010).<br />
Many plants are being used in the traditional medicine<br />
because they produce a diverse range of bioactive<br />
molecules, making them a rich source of different types<br />
of medicines (Tanaka et al., 2006). Dicranopteris linearis<br />
(Gleicheniaceae), known locally to the Malay’s as Resam<br />
has been used in the Malay’s traditional medicine as a<br />
cooling drink and also to reduce fever (Zakaria et al.,<br />
2006). In other part of the world, it is used to treat asthma<br />
and for women’s sterility (Vasuda, 1999), and to get rid of<br />
intestinal worms infection (Chin, 1992). Scientifically, D.<br />
linearis extracts have been reported to possess antinociceptive,<br />
anti-inflammatory and antipyretic activities<br />
(Zakaria et al., 2008), antibacterial activity (Lai et al.,<br />
2009) and potential cytotoxic and antioxidant activity<br />
against various types of cancer (Zakaria et al., 2011).<br />
Phytochemical study has revealed the presence of<br />
various types of flavonoids, particuarly of flavonol 3-Oglycosides<br />
types, and triterpenes, saponins and high<br />
content of steroids in the leaves of D. Linearis (Raja et<br />
al., 1995; Zakaria, 2007). Thus far, there is no data<br />
available on gastroprotective activity of DLELE. The<br />
present study was undertaken to evaluate antiulcerogenic<br />
properties of DLELE in rats.<br />
MATERIALS AND METHODS<br />
In this study, omeprazole was used as the reference anti-ulcer<br />
drug, and was obtained from the University Malaya Medical Centre<br />
(UMMC) Pharmacy. The drug was dissolved in carboxymethyl<br />
cellulose (0.5% w/v) (CMC) and administered orally to the rats in<br />
concentrations of 20 mg/kg body weight (5 ml/kg) according to the<br />
recommendation of (Abdulla et al., 2010).<br />
Plant specimen and extract preparation<br />
D. linearis leaves were obtained from Ethno Resources Sdn Bhd,<br />
Selangor Malaysia, and identified by comparison with the Voucher<br />
specimen deposited at the Herbarium of Rimba Ilmu, Institute of<br />
Science Biology, University of Malaya, Kuala Lumpur. The dried<br />
leaves were powdered using electrical blender. Hundred grams of<br />
the fine powder were soaked in 500 ml of 95% ethanol in conical<br />
flask for 3 days. After 3 days the mixture was filtered using a fine<br />
muslin cloth followed by filter paper (Whatman No. 1) and distilled<br />
under reduced pressure in an Eyela rotary evaporator (Sigma-<br />
Aldrich, USA). The dry extract was then dissolved in CMC(0.5%<br />
w/v) and administered orally to rats in concentrations of 250<br />
and 500 mg/kg body weight (5 ml/kg body weight) according to the<br />
recommendation of (Mahmood et al., 2010).<br />
Experimental animals for gastric ulcer<br />
Sprague Dawley healthy adult male rats were obtained from the<br />
Experimental Animal House, Faculty of Medicine, University of<br />
Malaya, and Ethic No. PM/27/07/2010/MAA (R). The rats were<br />
divided randomly into 4 groups of 6 rats each. Each rat that<br />
weighed between 200 - 225 g was placed individually in a separate<br />
cage (one rat per cage) with wide-mesh wire bottoms to prevent<br />
coprophagia during the experiment. The animals were maintained<br />
on standard pellet diet and tap water. The study was approved by<br />
the Ethics Committee for Animal Experimentation, Faculty of<br />
Medicine, University of Malaya, Malaysia. Throughout the<br />
experiments, all animals received human care according to the<br />
criteria outlined in the “Guide for the Care and Use of laboratory<br />
Animals” prepared by the National Academy of Sciences and<br />
published by the National Institute of Health.<br />
Gastric ulcer-induction by absolute ethanol<br />
The rats fasted for 48 h before the experiment (Abdulla et al.,<br />
2010), but were allowed free access to drinking water up till 2 h<br />
before the experiment. Gastric ulcer was induced by orogastric<br />
intubation of absolute ethanol (5 ml/kg) according to the method<br />
described by Mahmood et al. (2010). Ulcer control groups were<br />
orally administered vehicle (CMC, 0.5% w/v, 5 ml/kg). The<br />
reference group received oral doses of 20 mg/kg omeprazole in<br />
CMC (5 ml/kg) as positive control. Experimental groups were orally<br />
administered DLELE in CMC solution (5 ml/kg) at doses of 250 and<br />
500 mg/kg . One hour after this pre-treatment all groups of rats<br />
were administered with absolute ethanol (5 ml/kg) in order to induce<br />
gastric ulcers (Abdulla et al., 2010). The rats were euthanized 60<br />
min later (Ketuly et al., 2011) under an overdose of xylazin and<br />
ketamine anesthesia and their stomachs were immediately excised.<br />
Measurement of mucus production<br />
Gastric mucus production was measured in the rats that were<br />
subjected to absolute ethanol-induced gastric lesions. The gastric<br />
mucosa of each rat was obtained by gentle scraping the mucosa<br />
with a glass slide and the collected mucus were weighed by using a<br />
precision electronic balance (Ketuly et al., 2011; Wasman et al.,<br />
2010).<br />
Measurement of acid content of gastric juice (pH)<br />
Samples of gastric contents were analyzed for hydrogen ion<br />
concentration by pH metric titration with 0.1 N NaOH solutions<br />
using digital pH meter (Abdulla et al., 2010; Ketuly et al., 2011).<br />
Gross gastric lesions evaluation<br />
Ulcers of the gastric mucosa appear as elongated bands of<br />
hemorrhagic lesions parallel to the long axis of the stomach. Gastric<br />
mucosa of each rat was thus examined for damage. The length and<br />
width of the ulcer (mm) were measured by a planimeter (10 × 10<br />
mm 2 = ulcer area) under dissecting microscope (×1.8). The<br />
ulcerated area was measured by counting the number of small<br />
squares, 2 mm × 2 mm, covering the length and width of each ulcer<br />
band. The sum of the areas of all lesions for each stomach was
Table 1. Effect of DLELE on ulcer area and inhibition percentage in rats.<br />
Animal<br />
group<br />
Pre-treatment<br />
(5 ml/kg dose)<br />
Mucus<br />
production<br />
pH of gastric<br />
content<br />
Ulcer area (mm) 2<br />
(Mean ± S.E.M)<br />
Hussaini et al. 1763<br />
Inhibition<br />
(%)<br />
1 CMC (Ulcer control) 0.37 ± 0.01 a 3.88 ± 0.01 a 955.25 ± 2.82 a -<br />
2 Omeprazole (20 mg/kg) 0.57 ± 0.01 b 7.14 ± 0.33 b 148.3 � 2.62 b 84.48<br />
3 DLELE (250 mg/kg) 0.52 ± 0.01 c 5.6 ± 0.1 c 225.11 ± 3.65 c 76.43<br />
4 DLELE (500 mg/kg) 0.77 ± 0.01 d 6.22 ± 0.01 c 38.08 � 2.18 d 96.01<br />
All values are expressed as mean ± standard error mean. Means with different superscripts are significantly different. The mean<br />
difference is significant at the p>0.05 level.<br />
applied in the calculation of the ulcer area (UA) wherein the sum of<br />
small squares × 4 × 1.8 = UA (mm 2 ) according to the<br />
recommendation of Mahmood et al. (2010). The inhibition<br />
percentage (I.0%) was calculated using the following formula<br />
according to the recommendation of Wasman et al. (2010).<br />
(I%) = [(UAcontrol − UAtreated) � UAcontrol] × 100%.<br />
Histological evaluation of gastric lesions<br />
Specimens of the gastric walls of each rat were fixed in 10%<br />
buffered formalin and processed in a paraffin t<strong>issue</strong> processing<br />
machine. Sections of the stomach were made at a thickness of 5<br />
µm and stained with hematoxylin and eosin for histological<br />
evaluation (Abdulla et al., 2010; Ketuly et al., 2011).<br />
Statistical analysis<br />
All values were reported as mean ± S.E.M. Statistical significance<br />
of differences between groups was assessed using one-way<br />
ANOVA. A value of p
1764 Sci. Res. Essays<br />
Figure 1. Gross appearance of the gastric mucosa in rats. (a) pre-treated with 5 ml/kg CMC (ulcer<br />
control). Severe injuries are seen in the gastric mucosa (arrow). Absolute ethanol produced extensive<br />
visible hemorrhagic necrosis of gastric mucosa. (b) pre-treated with of omeprazole (20 mg/kg). Injuries to<br />
the gastric mucosa are very milder compared to the injuries seen in the ulcer control rats (arrow). (c) pretreated<br />
with DLELE (250 mg/kg). Mild injuries are seen in the gastric mucosa. The extract reduces the<br />
formation of gastric lesions induced by absolute ethanol (arrow). (d) pre-treated with DLELE 500 mg/kg.<br />
Mild injuries to the gastric mucosa are seen, and flattening of the gastric mucosa is seen (arrow).<br />
mucus production. This suggests that gastro-protective<br />
effect of DLELE is mediated partly by preservation of<br />
gastric mucus production.<br />
Oxidative stress plays an important role in the pathogenesis<br />
of various diseases including gastric ulcer, with<br />
antioxidants being reported to play a significant role in the<br />
protection of gastric mucosa against various necrotic<br />
agents (Trivedi and Rawal, 2001). Administration of<br />
antioxidants inhibits ethanol-induced gastric injury in rat<br />
(Ligumsky et al., 1995). DLELE possesses a broad<br />
spectrum of biological activities, and the plant extract has<br />
been shown to contain pharmaceutically active chemical<br />
constituents, such as flavonoids, saponins and<br />
terpenoids (Zakaria, 2007) and it is speculated that the<br />
gastroprotective effect exerted by DLELE could be<br />
attributed to its antioxidant property. Antioxidant property<br />
of the DLELE may possibly counteract oxidative damage<br />
caused by absolute ethanol toxicity. The observed<br />
anti-ulcerogenic activity may be due to its antioxidant<br />
effects and appears to strengthen the mucosal barrier,<br />
which is the first line of defense against endogenous and<br />
exogenous ulcerogenic agents. Previous studies have<br />
shown that flavonoids may be related to the antiulcer<br />
activity (Hiruma-Lima et al., 2006), and play a major role<br />
in the mechanism of gastroprotection (La Casa et al.,<br />
2000). It could be conceivable that the anti-ulcer activity<br />
of this plant could be linked to the flavonoids since<br />
flavonoids are reported to protect the mucosa by<br />
preventing the formation of lesions by various necrotic<br />
agents (Saurez et al., 1996). It is well known that many<br />
flavonoids display anti-secretory and cytoprotective<br />
properties in different experimental models of gastric<br />
ulcer (Zayachkivska et al., 2005). Flavonoids possess<br />
anti-oxidant properties in addition to strengthening the<br />
mucosal defense system through stimulation of gastric<br />
mucus secretion (Martin et al., 1994) and flavonoids can
Figure 2. Histological study of the absolute ethanol-induced gastric mucosal damage in rats (a) pretreated<br />
with 5 ml/kg of CMC (ulcer control). There is severe disruption to the surface epithelium,<br />
necrotic lesions penetrate deeply into mucosa (orang arrow) and extensive edema of submucosa<br />
layer and leucocyte infiltration are present (White arrow). (b) pre-treated with omeprazole (20<br />
mg/kg). Mild disruption of the surface epithelium mucosa are present but deep mucosal damage is<br />
absent. (c) pre-treated with DLELE (250 mg/kg). Moderate disruption of surface epithelium are<br />
present but deep mucosal damage is absent. There is edema and leucocytes infiltration of the<br />
submucosal layer. (d) pre-treated with DLELE (500 mg/kg). There is no disruption to the surface<br />
epithelium with no edema and no leuco.<br />
scavenge for the reactive oxygen species (super-oxide<br />
anions) and free radicals produced by ethanol. These<br />
reactive intermediates are potentially implicated in<br />
ulcerogenicity (Lewis and Hanson, 1991).<br />
It is generally known that the antioxidant activities of<br />
putative antioxidants involves various mechanisms, such<br />
as radical scavenging, decomposition of peroxides,<br />
binding of transition metal ion catalysts, prevention of<br />
chain initiation and of continued hydrogen abstraction<br />
(Diplock et al., 1998). Hence, the free radical scavenging<br />
capacity of an extract may serve as a significant indicator<br />
of its potential antioxidant activity. Increasing evidences<br />
have suggested that many age-related human diseases<br />
are the result of cellular damage caused by free radicals<br />
(Carr and Frei, 2000), Antioxidants have been shown to<br />
play an important role in preventing such diseases. For<br />
Hussaini et al. 1765<br />
example, several cancer chemopreventive agents exhibit<br />
antioxidant activity through their ability to scavenge<br />
oxygen radicals (Ito et al., 1999). DLELE demonstrated to<br />
contain flavonoids, saponins, triterpenes, tannins and<br />
steroids (Zakaria, 2007). The interests in phenolic<br />
compounds, particularly flavonoids and tannins, have<br />
considerably increased in recent years because of their<br />
broad spectrum of chemical and diverse biological<br />
properties, which include the antioxidant effects (Larson,<br />
1988) and radical scavenging properties (Agrawal, 1989).<br />
Flavonoids have been associated with possible role in the<br />
prevention of several chronic diseases involving oxidative<br />
stress (Lee et al., 2003), as well as their protective effect<br />
against low-density lipoprotein (LDL) oxidation (Silva et<br />
al., 2000). Flavonoids have been reported to inhibit<br />
cytokine (inflammatory stimuli) release from RAW264.7
1766 Sci. Res. Essays<br />
cells (Xagorari et al., 2002) and may modulate the<br />
increasing number of cellular processes involving redox<br />
reaction, including the regulation of tyrosine phosphatase<br />
activity (Gamet-Payrastre et al., 1999).<br />
The result of the present study also revealed protection<br />
of gastric mucosa and inhibition of leucocytes infiltration<br />
of gastric wall in rats pretreated with DLELE. DLELE<br />
have been shown to contain anti-inflammatory activity<br />
(Zakaria et al., 2006) and it is speculated that the<br />
gastroprotective effect exerted by this plant extract could<br />
be attributed to its anti-inflammatory activity. This antiinflammatory<br />
activity could also be a key factor in the<br />
prevention of gastric ulcer as reported by Swarnakar et<br />
al. (2005). Similarly, Abdulla et al. (2010) and Wasman et<br />
al. (2010) demonstrated that the reduction of neutrophil<br />
infiltration into ulcerated gastric t<strong>issue</strong> promotes the<br />
healing of gastric ulcers in rats. Mahmood et al. (2010)<br />
and Wasman et al. (2010) showed that oral administration<br />
of plant extract before ethanol administration<br />
significantly decreased neutrophil infiltration of gastric<br />
mucosa. Absolute alcohol would extensively damage the<br />
gastric mucosa leading to increased neutrophil infiltration<br />
into the gastric mucosa. Oxygenfree radicals derived<br />
from infiltrated neutrophils in ulcerated gastric t<strong>issue</strong>s<br />
have inhibitory effect on gastric ulcers healing in rats<br />
(Suzuki et al., 1998). Neutrophils mediate lipid peroxidation<br />
through the production of superoxide anions<br />
(Zimmerman et al., 1997). Neutrophils are a major source<br />
of inflammatory mediators and can release potent<br />
reactive oxygen species such as superoxide, hydrogen<br />
peroxide and myeloperoxidase derived oxidants. These<br />
reactive oxygen species are highly cytotoxic and can<br />
induce t<strong>issue</strong> damage (Cheng and Koo, 2000).<br />
Furthermore, neutrophil accumulation in gastric mucosa<br />
has been shown to induce gastric ulceration (Abdulla et<br />
al., 2010; Wasman et al., 2010; Ketuly et al., 2011).<br />
Suppression of neutrophil infiltration during inflammation<br />
was found to enhance gastric ulcer healing (Mahmood et<br />
al., 2010). Studies have demonstrated the link between<br />
the anti-inflammatory and antioxidant activities of the<br />
plants. For example, nitric oxide (NO) is produced/<br />
released under the action of inflammatory stimuli (that is,<br />
ROS) (Olszanecki et al., 2002). Inhibition of ROS leads to<br />
the reduction of NO production, which has been demonstrated<br />
to cause anti-inflammatory and antioxidant<br />
activities (Middleton et al., 2000). The free radical<br />
scavenging property may be one of the mechanisms by<br />
which these plants’ are effective in their ethnopharmacological<br />
uses against different ailments.<br />
In the present study, we observed flattening of the<br />
mucosal folds which suggests that gastroprotective effect<br />
of DLELE might be due to a decrease in gastric motility. It<br />
is reported that the changes in the gastric motility may<br />
play a role in the development and prevention of<br />
experimental gastric lesions (Abdulla et al., 2010; Ketuly<br />
et al., 2011). Relaxation of circular muscles may protect<br />
the gastric mucosa through flattening of the folds. This<br />
will increase the mucosal area exposed to necrotizing<br />
agents and reduce the volume of the gastric irritants on<br />
rugal crest (Mahmood et al., 2010; Wasman et al., 2010).<br />
Ethanol produces a marked contraction of the circular<br />
muscles of rat fundic strip. Such a contraction can lead to<br />
mucosal compression at the site of the greatest<br />
mechanical stress, at the crests of mucosal folds leading<br />
to necrosis and ulceration (Abdulla et al., 2010).<br />
Conclusion<br />
The study reveals DLELE could significantly protect the<br />
gastric mucosa against ethanol-induced injury. Such<br />
protection was ascertained grossly by increased gastric<br />
mucus production, decrease in the acidity of gastric<br />
content were significantly higher in treated groups<br />
compare to ulcer control group and also the reduction of<br />
ulcer areas in the gastric wall as well as histology by the<br />
reduction or inhibition of edema and leucocytes infiltration<br />
of submucosal layers. The data obtained confirm the<br />
traditional indications for this herb and present a new<br />
therapeutic option for the treatment of gastric ailments.<br />
The exact mechanism (s) underlying this anti-ulcerogenic<br />
effect remain unknown, but it seems that this extract<br />
contains pharmacologically active substances with potent<br />
antioxidant and anti-inflammatory activity which increase<br />
the mucus production and decrease the acidity of gastric<br />
content.<br />
ACKNOWLEDGEMENTS<br />
This study was financially supported by the University of<br />
Malaya through University Malaya Research Grant 2009<br />
(UMRG), RG102/09HTM and 600-RMI/ERGS 5/3<br />
(26/2011).<br />
REFERENCES<br />
Abdulla MA, Ahmed KAA, AL-Bayaty, FH, Masood, Y (2010).<br />
Gastroprotective effect of Phyllanthus niruri leaf extract against<br />
ethanol-induced gastric mucosal injury in rats. Afr. J. Pharm.<br />
Pharmacol., 4(5): 226-230.<br />
Agrawal PK (1989). Carbon-13 NMR of Flavonoids. New York: Elsevier<br />
Press, pp. 5-31.<br />
Al-Mofleh IA, Alhaider AA, Mossa JS, Al-Soohaibani MO, Rafatullah S<br />
(2007). Aqueous suspension of anise Pimpinella anisum protects rats<br />
against chemically induced gastric ulcers. World J. Gastroenterol.,<br />
13: 1112-1118.<br />
Carr A, Frei B (2000). The role of natural antioxidants in preserving the<br />
biological activity of endothelium-derived nitric oxide. Free Rad. Biol.<br />
Med., 28: 1806-1814.<br />
Cheng CL, Koo MWL (2000). Effect of Centella asiatica on ethanol<br />
induced gastric mucosal lesions in rats. Life Sci., 67: 2647-2653.<br />
Chin WY (1992). A Guide to Medicinal Plants. Singapore Sci. Centre, p.<br />
24.<br />
Diplock AT, Charleux JL, Crozier-Willi G, Kok FJ, Rice-Evans C,<br />
Roberfroid M, Stahl W, Vina-Ribes J (1998). Functional food science<br />
and defence against reactive oxygen species. Br. J. Nutr., 80: S77-<br />
S112.
Gamet-Payrastre L, Manenti S, Gratacap MP, Tulliez J, Chap H,<br />
Payrastre B (1999). Flavonoids and the inhibition of PKC and PI 3-<br />
kinase. Gen. Pharmacol., 32: 279-286.<br />
Hiruma-Lima CA, Calvo TR, Rodrigues CM, Andrade FD, Vilegas W,<br />
Brito AR (2006). Antiulcerogenic activity of Alchornea castaneafolia:<br />
effects on somatostatin, gastrin and prostaglandin. J.<br />
Ethnopharmacol., 104: 215-224.<br />
Ito H, Miyake M, Nishitani E, Mori K, Hatano T, Okuda T, Konoshima T,<br />
Takasaki M, Kozuka M, Mukainaka M, Tokudo H, Nishino H, Yoshida<br />
T (1999). Antitumor promoting activity of polyphenols from Cowania<br />
mexicana and Coleogyne ramosissima. Cancer Lett., 143: 5-13.<br />
Khazaei M, Salehi H (2006). Protective effect of falcaria vulgaris extract<br />
on ethanol induced gastric ulcer in rat. Iran J. Pharmacol. Ther., 5:<br />
43-46.<br />
Ketuly KA, Abdulla MA, Hadi HA, Mariod AA, Abdel-Wahab SI (2011).<br />
Anti-ulcer activity of the 9alpha-bromo analogue of Beclomethasone<br />
dipropionate against ethanol-induced gastric mucosal injury in rats. J.<br />
Med. Plants Res., 5(4): 514-520.<br />
La Casa C, Villegas I, Alarcon de la Lastra C, Motilva V, Martin Calero<br />
MJ (2000). Evidence for protective and antioxidant properties in rutin,<br />
a natural flavones, against ethanol induced gastric lesions. J.<br />
Ethnopharmacol., 71: 45-53.<br />
Lai HY, Lim YY, Tan SP (2009). Antioxidative, tyrosinase inhibiting and<br />
antibacterial activities of leaf extract from medicinal ferns. Biosci.<br />
Biotechnol. Biochem., 73(6): 1362-1366.<br />
Larson RA (1988). The antioxidants of higher plants. Phytochem., 27:<br />
969-978.<br />
Lee KW, Kim YJ, Lee HJ, Lee CY (2003). Cocoa has more phenolic<br />
phytochemicals and a higher antioxidant capacity than teas and red<br />
wine. J. Agric. Food Chem., 51: 7292-7295.<br />
Lewis DA, Hanson PJ (1991). Antiulcer drugs of plant origin. Prog. Med.<br />
Chem., 28: 210-229.<br />
Ligumsky M, Sestieri M, Okon F, Ginsburg I (1995). Antioxidants inhibit<br />
ethanol-induced gastric injury in the rat. Role of manganese, glycin<br />
and carotene. Scand. J. Gastroenterol., 30: 854-860.<br />
Olszanecki R, Gêbska A, Kozlovski VI, Grylewski RJ (2002). Flavonoids<br />
and nitric oxide synthase. J. Physiol. Pharmacol., 53: 571-584.<br />
Mahmood AA, Mariod AA, Al-Bayaty F, Abdel-Wahab SI (2010). Antiulcerogenic<br />
activity of Gynura procumbens leaf extract against<br />
experimentally-induced gastric lesions in rats. J. Med. Plants Res.,<br />
4(8): 685-691.<br />
Martin MJ, Marhuenda E, Perez-Guerrero C, Franco JM (1994).<br />
Antiulcer effect of naringin on gastric lesion induced by ethanol in<br />
rats. Pharmacology, 49: 144-150.<br />
Middleton Jr. E, Kandaswami C, Theoharides TC (2000). The effects of<br />
plant flavonoids on mammalian cells: implications for inflammation,<br />
heart disease, and cancer. Pharmacol. Rev., 52: 673-751.<br />
Mizui T, Sato H, Hirose F, Doteuchi M (1987). Effect of antiperoxidative<br />
drugs on gastric damage induced by ethanol in rats. Life Sci., 41:<br />
755-763.<br />
Raja DP, Manickam VS, De Britto AJ, Gopalakrishnan S, Ushioda T,<br />
Satoh M, Tanimura A, Fuchino H, Tanaka N (1995). Chemical and<br />
chemotaxonomical studies on Dicranopteris species. Chem. Pharm.<br />
Bull., 43: 1800-1803.<br />
Saurez J, Herreta MD, Marhuenda E (1996). Hesperidine and<br />
neohesperidine dihydrochalcone on different experimental models of<br />
induced gastric ulcer. Phytother. Res., 10: 616-618.<br />
Hussaini et al. 1767<br />
Silva FA, Borges F, Guimaraes C, Lima JL, Matos C, Reis S (2000).<br />
Phenolic acids and derivatives: Studies on the relationship among<br />
structure, radical activity, and physicochemical parameters. J. Agric.<br />
Food Chem., 48: 2122-2126.<br />
Suzuki Y, Ishihara M, Ito M (1998). Anti-ulcer effects of antioxidants,<br />
quercetin, �-tocopherol, nifedipine and tetracycline in rats. Jpn. J.<br />
Pharmacol., 78: 435-441.<br />
Swarnakar S, Ganguly K, Kundu P, Banerjee A, Maity P, Sharma AV<br />
(2005). Curcumin regulates expression and activity of matrix<br />
metalloprotinases 9 and 2 during prevention and healing of<br />
indomethacin-induced gastric ulcer. J. Biol. Chem., 280: 9409-9415.<br />
Tanaka JCA, Da Silva CC, De Oliveira AJB, Nakamura CV, Dias Filho<br />
BP (2006) Antibacterial activity of indole alkaloids from Aspidosperma<br />
ramiflorum. Braz. J. Med. Biol. Res., 39: 387-391<br />
Trivedi NP, Rawal UM (2001). Hepatoprotective and antioxidant<br />
property of A. paniculata Nees in BHC induced liver damage in mice.<br />
Indian J. Exp. Biol., 39: 41-46.<br />
Vasuda SM (1999). Economic importance of pteridophytes. Indian Fern.<br />
J. 16: 130-152.<br />
Wasman SQ, Mahmood AA, Salehhuddin H, Zahra AA, Salmah I<br />
(2010). Cytoprotective activities of Polygonum minus aqueous leaf<br />
extract on ethanol-induced gastric ulcer in rats. J. Med. Plants Res.,<br />
4(24): 2658-2665.<br />
Xagorari A, Roussos C, Papapetrpoulos A (2002). Inhibition of<br />
LPSstimulated pathways in macrophages by the flavonoid luteolin.<br />
Br. J. Pharmacol., 136: 1058-1064.<br />
Zakaria ZA (2007). Free radical scavenging activity of some plants<br />
available in Malaysia. Iran J. Pharmacol. Ther., 6: 87-91.<br />
Zakaria ZA, Abdul Ghani ZDF, Raden Mohd Nor RNS, Hanan Kumar G,<br />
Sulaiman MR, Fatimah CA (2006). Antinociceptive and<br />
antiinflammatory activities of Dicranopteris linearis leaves chloroform<br />
extract in experimental animals. Yakugaku Zasshi, 126: 1197-1203.<br />
Zakaria ZA, Ghani ZD, Nor RN, Gopalan HK, Sulaiman MR, Jais AM,<br />
Somchit MN, Kader AA, Ripin J (2008). Antinociceptive, antiinflammatory,<br />
and antipyretic properties of an aqueous extract of<br />
Dicranopteris linearis leaves in experimental animal models. J. Nat.<br />
Med., 62(2):179-87.<br />
Zakaria ZA, Mohamed AM, Mohd. Jamil NS, Rofiee MS, Somchit MN,<br />
Zuraini A, Arifah AK, Sulaiman MR (2011). In vitro cytotoxic and<br />
antioxidant properties of the aqueous, chloroform and methanol<br />
extracts of Dicranopteris linearis leaves. Afr. J. Biotechnol., 10(2):<br />
273-282.<br />
Zayachkivska OS, Konturek SJ, Drozdowicz D, Konturek PC,<br />
Brzozowski T, Ghegotsky MR (2005). Gastroprotective effects of<br />
flavonoids in plant extracts. J. Physiol. Pharmacol., 56: 219-231.<br />
Zimmerman JJ, Ciesielski W, Lewandoski J (1997). Neutrophilmediated<br />
phospholipids peroxidation assessed by gas<br />
chromatography-mass spectroscopy. Am. J. Physiol., 273: 653-661.
Scientific Research and Essays Vol. 7(18), pp. 1768-1776, 16 May, 2012<br />
Available online at http://www.academicjournals.org/SRE<br />
DOI: 10.5897/SRE11.1483<br />
ISSN 1992-2248 © 2012 <strong>Academic</strong> <strong>Journals</strong><br />
Full Length Research Paper<br />
Introducing slag powder as drag reduction agent in<br />
pipeline: An experimental approach<br />
Hayder A. Abdulbari*, Siti Nuraffini Bt Kamarulizam, Rosli M. Y. and Arun Gupta<br />
Faculty of Chemical and Natural Resources Engineering, University Malaysia Pahang, Lebuhraya Tun Razak, 26300<br />
Kuantan, Pahang Darul Makmur, Malaysia.<br />
Accepted 20 September, 2011<br />
Main by product in ore smelting from tin production in Malaysia has become a trigger for this<br />
investigation. Slag waste can be categorized as suspended solid. Utilization of this waste in fluid<br />
transportation can reduce the pressure drop in pipelines. Experimental works had conducted in order to<br />
test slag waste in a closed loop of turbulence water flowing system with water and fuel as the transport<br />
liquid. The procedures start by pumping liquid suspended solid combination from reservoir tank with<br />
varies flow rates into two different pipe diameters (0.0127 m ID and 0.038 m D.I). The types of pipe used<br />
are PVC pipe. The testing length of this flow system is 2.0 m. The pressure drop and drag reduction<br />
were measured in varied addition concentration. The results have show percentage drag reduction<br />
(Dr%) is over 60% in certain range and condition. It is proved that slag is a potential DRA.<br />
Key words: Suspended solid, turbulent flow, drag reducing agent, pipeline system.<br />
INTRODUCTION<br />
Drag reduction in turbulent flow is a very important<br />
subject in technologies utilization and significant point of<br />
interest. As known, drag reduction can be achieved by<br />
using several numbers of additives that have been widely<br />
studied such as high molecular weight polymers (Roy<br />
and Larson, 2005; Al-Sarkhi, 2010, Mowla and Naderi,<br />
2006; Shetty and Solomon, 2009; Parimal et al., 2008;<br />
Janosi et al., 2004; Dubief et al., 2004) surfactants study<br />
by Lu et al. (1998), Bari and Yunus (2009), Bari et al.<br />
(2008) and Suali et al. (2010) and suspended solid<br />
investigation have been determined by Roy and Larson<br />
(2005) and Dyer et al. (2004). Drag reduction is defined<br />
as addition of several ppm concentrations to accelerate<br />
radically in fluid transportation (Brostow et al., 2006; Suali<br />
et al., 2010; Dubief et al., 2004; Kim et al., 2000; Parimal<br />
et al., 2008) and many more. However, there are few<br />
studies regarding suspended solid. Mechanism involved<br />
is yet still a hesitation. Many researchers investigated the<br />
idea of particles inside liquid flow channel agreed in one<br />
point, turbulent characteristic are changed in present<br />
*Corresponding author. E-mail: hayder.bari@gmail.com. Tel:<br />
0060123495130.<br />
of suspended solid (Rashidi et al., 1990; Roy and Larson,<br />
2005; Dyer et al., 2004; Filipson et al., 1977). There are<br />
fewer studies devoted to the effects of particle additives<br />
on the mechanisms of instability and transition to<br />
turbulence in free shear flows. The flow visualizations<br />
reported by Filipsson et al. (1977) represent one of the<br />
few available experiments on this subject. In this study,<br />
the authors presented results for a jet flow of viscoelastic<br />
(Polyox WSR-301), fibre suspension (chrysotile fibres)<br />
and Newtonian (water) fluids at high Reynolds numbers.<br />
Rashidi et al. (1990) have determined predominant<br />
effects of particle size, density and concentration. Their<br />
results point out that particle density has minor effects<br />
rather than particle size in effects of drag reduction.<br />
Presently, the usage of the suspended solid (insoluble<br />
in liquid media) as drag reducing agent has broaden<br />
accesses for enthusiastic researcher to uncover the<br />
accessibility of this insoluble condition in the drag<br />
reduction phenomena (Toonder, 1997; Mowla and<br />
Naderi, 2006). The aim of this study is to test the<br />
efficiency of slag particle as drag reducer agent on<br />
transport of fuel inside pipes. Two different internal pipe<br />
diameters were used with four different concentrations in<br />
the purpose to investigate the concentration effect. The<br />
efficiency of suspended solid was tested using diesel.
Instrument<br />
P1 – P5 Pressure sensor for 0.0381 m D.I<br />
P6 – P10 Pressure sensor for 0.0254 m D.I<br />
Valve<br />
Centrifugal pump<br />
Figure 1. Schematic diagram of the flow system.<br />
MATERIALS AND METHODS<br />
Liquid circulation system<br />
Liquid circulation system was built to test the effects of pipe<br />
diameter, pipe length, fluid velocity and concentration on pressure<br />
drop; hence to investigate the effects influence %Dr. Figure 1<br />
shows a schematic diagram of a build up liquid circulation system<br />
used in the present investigation. Generally, this system consists of<br />
reservoir tank, pipes, valves, pumps, flow meter and pressure<br />
sensors. The reservoir tank was supported with an exit pipe<br />
connected to centrifugal pumps. The solution flows from reservoir<br />
tank directly into testing pipes before flowing back into the reservoir<br />
tank. Two visible PVC pipes connected to galvanized iron pipes of<br />
various inside diameters 0.0127 and 0.0381 m ID were used in<br />
constructing the flow system. A <strong>complete</strong> closed loop piping system<br />
Abdulbari et al. 1769<br />
was built. Flow starts from the reservoir tank through the pump<br />
reaching a split connected with three different pipes diameter with<br />
testing section. For measuring flow rate inside the pipe, Burkert<br />
attachable flow meter has been used. Power supply for this<br />
detector is 12 to 30 V and can measure flow pressure between 140<br />
to 230 psi with fluid viscosity les then 200 cP. Material used to be<br />
immersed into fluid is ceramic. The pump brand for water circulation<br />
is Grundfos CH8-40. The energy usage of this pump is 1.02 kW<br />
and pump force is 415 V. On the other hand, hydrocarbon pump<br />
brand is DAB K-Series which is transfer pump for small flow<br />
operations. This centrifugal pump is designed with technopoymer<br />
impeller for domestic usage. The energy required for this pump is<br />
0.75 kW and pump force is 240 V. The testing sections were 0.5,<br />
1.0, 1.5 and 2.0 m long. First, testing point was located about 50<br />
times of pipe diameter to ensure that the turbulent flows are fully<br />
developed before the testing process run. Five sets of build up
1770 Sci. Res. Essays<br />
(a) (b)<br />
Figure 2. Grinded slag waste particle for 200 μm: (a) SEM picture of 9x mag and (b) Scaled up picture of a particle<br />
(33x mag).<br />
Table 1. Physical properties of diesel.<br />
Hydrocarbon properties at 27°C<br />
Viscosity (μ diesel at 27°C ) 1456 cP<br />
Density (ρ diesel at 27°C ) 853.2 kg/m 3<br />
Table 2. Physical properties of water.<br />
Water properties at 27°C<br />
Viscosity (μ water at 27°C ) 937.3 cP<br />
Density (ρ waterat 27°C ) 996.59 kg/m 3<br />
pressure sensors.<br />
The pressure sensor is a build up pressure transmitter and<br />
transducer made by silicon as the base materials. The membrane<br />
detector is made by 316 L stainless steel diaphragm. The optimum<br />
range for pressure is 0 to 60 bar, temperature range between -10<br />
until 60°C and be detected in absolute gauge. The accuracy for this<br />
transmitter is 0.1% fault. These sensors are integrated construction<br />
specially design for rigid and robust flow in industry.<br />
Materials investigated (slag waste particle)<br />
Slag waste is obtained from Malaysia Smelting Corporation at<br />
Butterworth, Penang. The slag waste was dried by the oven<br />
overnight at a temperature of 100°C. Once dry, the suspended solid<br />
is graded into fine particle by using grinder then sample was sieve<br />
using a screen into 200 μm in size (Figure 2). The density of this<br />
new material is 1400 kg/m 3 .<br />
Transported liquid<br />
The transported liquid used in the present investigation was diesel<br />
fuel obtained from Shell. However, for comparison, pure water was<br />
used. The physical properties of diesel fuel and water are shown in<br />
Tables 1 and 2.<br />
Experimental procedure<br />
All the experiments were carried in a constructed liquid circulation<br />
system, testing different variables which are:<br />
i) Suspended solid concentration (50, 100, 200 and 400 ppm).<br />
ii) Pipe diameter (0.0127 and 0.0381 m D.I).<br />
iii) Pipe length (0.5, 1.0, 1.5 and 2.0 m).<br />
iv) Solution flow rates.
Abdulbari et al. 1771<br />
Figure 3. Effect of particle concentration in transported fuel on Dr% with different Re in 0.0381 n D.I and 2.0 m pipe length<br />
and pressure drop analysis for Re = 105000.<br />
The experimental procedure starts by testing every additive<br />
concentration and pipe diameter, the operation begins when the<br />
pump starts delivering the solution through section of pipe length.<br />
The solution flow rate is fixed at the certain value by controlling it<br />
from the bypass section. Pressure readings are taken to this flow<br />
rate. By changing the solution flow rate to another fixed point,<br />
pressure readings are taken again until finishing desired values of<br />
flow rates. This procedure is repeated for each suspended solid<br />
concentrations to test its effect on the drag reduction operation.<br />
From the flow rate obtain, Reynolds number calculated based on<br />
formulae, which represent velocity of flow depending on pipe<br />
diameter and pipe length.<br />
Velocity and Reynolds number calculations<br />
The average velocity (V) and Reynolds number (Re) were<br />
calculated using the solution volumetric flow rate readings (Q),<br />
density (ρ), viscosity (μ) and pipe diameter (D) for each run as<br />
follows:<br />
�V. . D<br />
Re�<br />
�<br />
Where:<br />
ρ is density of the fluid (kg/m 3 ),<br />
µ is viscosity of the fluid (Pa. s),<br />
D is diameter of internal pipe (m),<br />
V is velocity of fluid (m/h),<br />
Where: =<br />
(1)<br />
Where: =<br />
After obtaining Reynolds number, percentage drag reduction (%Dr)<br />
determined to plot graft in order to see the patent of drag reduction.<br />
Percentage drag reduction calculations<br />
Pressure drop readings through testing sections before and after<br />
drag reducer addition were needed to calculate the percentage<br />
drag reduction %Dr as follows (Virk, 1967):<br />
Pb<br />
Pa<br />
Pb<br />
% Dr��<br />
�<br />
Where:<br />
� �<br />
∆Pb = Pressure drop before addition of DRA.<br />
∆Pa = Pressure drop after addition of DRA.<br />
RESULTS<br />
The analysis in Figures 3 and 4 shows slag particle<br />
performance on drag reduction as a function of particle<br />
concentration in hydrocarbon liquid and water and fluid<br />
velocity represented by Re which is from 50 to 400 ppm<br />
in the range of Re equal to 65000 to 140000 for fuel and<br />
65000 to 105000 for water transportation. Scale down for<br />
(2)
1772 Sci. Res. Essays<br />
Figure 4. Effect of particle concentration in transported water on Dr% with different Re in 0.0381 n D.I and 2.0 m pipe length and<br />
pressure drop analysis for Re = 105000.<br />
Figure 5. Effect of pipe diameter flowing through 2 m pipe length in fuel and 400 ppm concentration addition and pressure drop<br />
data for pipe diameter of 0.0127 and 0.0381 m D.I.<br />
one Re showed different pressure drop from raw data<br />
obtained as seen in Figures 3 and 4 for Re = 14000 for<br />
fuel and 105000 for water. The analysis in Figures 5 and<br />
6 shows slag particle performance on drag reduction as a<br />
function of pipe diameter in hydrocarbon liquid and water<br />
and fluid velocity represented by Re which is for 400 ppm<br />
in the range of Re equals 45000 to 145000 for fuel and<br />
Re equals 32000 to 95000 for water. Scale down for one<br />
Re, the different of pressure drop in two different pipe<br />
diameters. For hydrocarbon, Re for raw pressure drop<br />
data taken at 105000 for 0.0127 m D.I and 145000 for<br />
0.0381 m D.I. For water, Re for raw pressure drop data
Abdulbari et al. 1773<br />
Figure 6. Effect of pipe diameter flowing through 2 m pipe length in water and 400 ppm concentration addition and pressure<br />
drop data for pipe diameter of 0.0127 and 0.0381 m D.I.<br />
Figure 7. Effect of pipe length in 400 ppm addition concentration flowing in fuel with pipe diameter 0.0381 m D.I and pressure drop<br />
data for Re = 139000.<br />
taken at 70000 for 0.0127 m D.I and 95000 for 0.0381 m<br />
D.I. The analysis in Figures 7 and 8 shows slag particle<br />
performance on drag reduction as a function of pipe<br />
length in hydrocarbon liquid and water and fluid velocity<br />
represented by Re which is for 400 ppm in the range of<br />
Re equal to 98000 to 145000 for fuel and Re equal 65000<br />
to 105000 for water. Scale down for one Re, the different<br />
of pressure drop in four different testing sections which
1774 Sci. Res. Essays<br />
Figure 8. Effect of pipe length in 400 ppm addition concentration flowing in water with pipe diameter 0.0381 m D.I and<br />
pressure drop data for Re = 100000.<br />
are 0.5, 1.0, 1.5 and 2.0 m. For hydrocarbon, Re taken at<br />
point equal to 139000 and for water, Re taken at point<br />
equal to 100000.<br />
DISCUSSION<br />
Note that the profile pattern of each concentration and<br />
fluid velocity is similar but varies in its value. Both result<br />
shows, either in fuel or water, drag reduction is increased<br />
by increasing of particle addition concentration. These<br />
data complied with many other researchers such as<br />
Rashidi et al. (1990), Roy and Larson (2005), Dyer et al.<br />
(2004) and Filipson et al. (1977). There are two result<br />
which shows drag reducer performance towards each<br />
solvent. These results showed that the optimum<br />
performances of suspended solid additive are limited to<br />
the degree of turbulence. The increment of fluid velocity<br />
in flow will establish a degree of turbulences in its own<br />
fluid range. However, there are two explanations that can<br />
be made for these two results. For Figure 3, drag<br />
reduction efficiency are decreased by increasing<br />
Reynolds number. However, drag reduction are still<br />
recorded in the flow. For different type of solvent,<br />
efficiency for same drag reducer are different. Due to<br />
lighter fuel density and viscosity compare to water, slag<br />
particle seem too ‘heavy’ to be pushed; so that, along<br />
pipe flow, the particle are deposited at the bottom of pipe<br />
wall little by little and result to few particle to stay flow<br />
with fluid. These phenomena have been explained by<br />
Dyer et al. (2004). Figure 4 shows higher fluid velocity<br />
gives higher degree of turbulent which will provide more<br />
suitable environment or possibility for the drag reduction<br />
mechanism to perform. Further increase can happen until<br />
it comes to the limitation. Over its degree of turbulent will<br />
cause the reduction static due to decrement of additive<br />
efficiency. The reason for these phenomena is because<br />
of the impairing ratio of additive concentration and degree<br />
of turbulence.<br />
Figures 5 and 6 shows the same pattern of drag<br />
reduction in function of additive concentration and fluid<br />
velocity. However, the value in 0.0127 m D.I is obviously<br />
smaller compared to 0.0381 m D.I. Increasing the pipe<br />
diameter means increasing the velocity inside the pipe in<br />
range of same flow rate resulting in increment of<br />
turbulence. The presence of eddies means that the local<br />
velocity is not the same as the bulk velocity and that<br />
there are components of velocity in all directions. The<br />
result complied with Bari et al. (2008). In larger pipe, the<br />
energy in the pipe is larger than smaller pipe due to fluid<br />
quantity itself. Higher energy is maintained inside large<br />
pipe rather than small pipe because of less different of<br />
local and bulk velocity hence decrease shear stress<br />
formation. However, in smaller pipes, high pressures are<br />
obtained because of fluid flow from large area to small<br />
area. When pressure is high, shear stress between local<br />
velocity and bulk velocity is significantly, resulting to more<br />
amount of energy absorbed to form eddies. However,<br />
due to higher volume of fluid, higher degree of turbulence<br />
are obtained in larger pipes because of higher Reynolds<br />
number resulting to higher possible collisions between<br />
eddies and more space of DRA to take action. These<br />
collisions provide extra number of eddies absorbing<br />
energy from the main flow to <strong>complete</strong> their shape. By
addition of DRA, eddies is bursting and pressure can be<br />
maintained along the pipes. However, pressure are not at<br />
the same maintaining state in order to burst eddies along<br />
the pipes. Also include friction of particle itself with pipe<br />
walls will reduce the efficiency. This can be proofed by<br />
Figures 5 and 6. Both figure shows by increasing pipe<br />
length, most of %DR is decreased slightly after<br />
maintaining the reduction for a while. For additional<br />
information, particle diameter effect towards slag particle<br />
flexibility to act as drag reducer. In order to be a good<br />
drag reducer, basic criteria such as flexibility and surface<br />
roughness should be fulfilled (Singh, 1990). However,<br />
fluctuation in drag reduction reading is due to high<br />
turbulent intensities which mean that the flow is very<br />
unstable. This phenomena happen in all variables<br />
(concentrations, time, pipe length and pipe diameter) as<br />
can be seen in Figures 3, 4, 5, 6, 7 and 8 for both solvent<br />
conditions.<br />
From Figures 7 and 8, drag reduction have occur in first<br />
section which means higher drag reduction and the<br />
effectiveness lesser within the length due to less<br />
momentum inside these particles. As known, slag particle<br />
cannot be degraded so that when particles are ‘pushed’<br />
by instable turbulent flow; it then increased the<br />
momentum back and performs drag reduction again.<br />
These processes will be repeated along the pipe as long<br />
as these particle remains inside the flow and not be<br />
degraded by any condition. For the industrial activities,<br />
longer pipes and bigger diameters are vital in long<br />
distance transportation. The patterns that drag reduction<br />
have proved can be considered as potential DRA. Due to<br />
very stable particles, the drag reduction will give up and<br />
down result as explained earlier. However, these<br />
particles are maintained to give effect on drag reduction<br />
along pipe length without addition of extra DRA. Bigger<br />
pipe also give advantages to this DRA since more space<br />
for DRA mechanism to perform and give less turbulent<br />
intensities hence increased drag reduction efficiency. As<br />
proved in these graphs data and present results, there<br />
are some weaknesses in using present DRA. Time<br />
consumption is needed in order to separate DRA with<br />
transported solvent. The separation begins by<br />
sedimentation of DRA to bottom of tank and transported<br />
solvent can be flow gravitational or by pump to other<br />
section for further use. However, because this particle is<br />
inert and stable which do not react with transported liquid,<br />
letting DRA immersed for a while is not a big problem.<br />
These present DRA also give certain advantages<br />
compared to polymeric DRA, surfactant DRA and fibers<br />
DRA. Compared to other DRA, there were no reinsertions<br />
of DRA once added into transported liquid. This will<br />
reduce operation cost. By using present DRA, the<br />
characteristic and properties of transported liquid will not<br />
ever be disturbed. Compare to polymer, degradation is a<br />
major problem since polymer chain can broke due to<br />
mechanical force (pump rotation) and biodegradable over<br />
time. For surfactant, micelle formation due to shear stress<br />
Abdulbari et al. 1775<br />
will reduce drag reduction efficiency.<br />
In hydrocarbon, surfactant can trap water to move<br />
along and will give emulsion. This emulsion is a great<br />
problem since to separate required much more time and<br />
temperature control. Fibre which categorized as<br />
suspended solid also is not very suitable in transporting<br />
hydrocarbon since it can change the hydrocarbon<br />
characteristic.<br />
Conclusion<br />
It is concluded that slag waste particle is applicable as<br />
suspended solid DRA since it is economic, inert (do not<br />
react with transported fluid), effective in water and diesel,<br />
potentially in refinery products. Several effects have been<br />
investigated in order to this new economic solver and<br />
how the drag reducing work. It had proved that slag<br />
particle will increase %Dr by increasing fluid velocity and<br />
concentration and pipe diameter. There is several<br />
limitation of using this suspended solid. Suspended solid<br />
can be effective in certain range degree of turbulent but<br />
will give sedimentation at the base of storage tank. As for<br />
recommendation, stirrer can be installed inside tank to<br />
ensure particles are evenly distributed. This suspended<br />
solid is not harmful to living organism and environment<br />
since there is no hazardous chemical that have been<br />
used.<br />
ACKNOWLEDGEMENTS<br />
Deepest gratitude is expressed to Universiti Malaysia<br />
Pahang and Faculty of Chemical and Natural Resources<br />
Engineering for providing the grant and facilities to<br />
support this research work. Appreciation extended to the<br />
supervisor, co-supervisor and laboratory staff for their<br />
guidance, advice, support and encouragements.<br />
REFERENCES<br />
Abdul Bari HA, Mohd RY (2009). Sodium Stearate as Drag Reducing<br />
Agent in Non-Aqueous Media. Int. J. Chem. Tech., 1(1): 11-18.<br />
Al-Sarkhi A (2010). Drag reduction with polymers in gas-liquid/ liquidliquid<br />
flows in pipes: A Literature Review. J. Nat. Gas Sci. Eng., 2:<br />
41-48.<br />
Bari HAA, Suali E, Hassan Z (2008). Glycolic acid ethoxylate lauryl<br />
ether performances as drag reducing agent in aqueos media flow in<br />
pipelines. J. Appl. Sci., 8: 4410-4415.<br />
Brostow W, Haley EHL, Reddy T, Singh RP, White L (2006). Lowering<br />
mechanical degradation of drag reducers in turbulent flow. J. Mater.<br />
Res., 22(1): 56-60.<br />
Dubief Y, White CM, Terrapon VE, Shaqfeh ESG, Moin P, Lele SK<br />
(2004). On the coherent drag-reducing and turbulence-enhancing<br />
behavior of polymers in wall flows. J. Fluid Mech., 514: 271-280.<br />
Dyer KR, Christie MC, Manning AJ (2004). The effects of suspended<br />
sediment on turbulence within an estuarine tirbudity maximum. J. Est.<br />
Coast Shelf Sci., 59: 237-248.<br />
Filipsson LGR, Torgny Lagerstedt JH, Bark FH (1977). A note on the<br />
analogous behavior of turbulent jets of dilute surfactant solutions and<br />
fibre suspensions. J. non-Newt Fluid Mech., 3: 97-103.
1776 Sci. Res. Essays<br />
Janosi IM, Jan D, Szabó KG, Tel T (2004). Turbulent drag reduction in<br />
dam-break flows. Exp. Fluids., 37: 219-229.<br />
Kim CA, Kim JT, Lee K, Choi HJ, Jhon MS (2000). Mechanical<br />
degradation of dilute polymer solutions under turbulent flow, J. Pol.,<br />
41: 7611-7615.<br />
Lu B, Li X, Scriven le, Davis HT, Talmon Y, Zakin JL (1998). Effect of<br />
Chemical Structure on Viscoelasticity and Extensional Viscosity of<br />
Drag-Reducing Cationic Surfactant Solutions. J. Chem. Struct.<br />
Visco, Langmuir., 14(1): 8-16.<br />
Mowla D, Naderi A (2006). Experimental study of drag reduction by a<br />
polymeric additive in slug two-phase flow of crude oil and air in<br />
horizontal pipes. Chem. Eng. Sci., 61: 1549-1554.<br />
Parimal PM, Cheolho K, Alvaro AOM (2008). The Performance of Drag<br />
Reducing Agent in Multiphase Flow Conditions at High Pressure;<br />
Positive and Negatives Effects. Proc. Of 7th Int. Pipeline Conf.,<br />
Calgary, Alberta, Canada.<br />
Rashidi M, Hetsroni G, Banerjee S (1990). Particle-turbulence<br />
interaction in a boundary layer. Int. J. Mult. Flow, 16(6): 935-949.<br />
Roy A, Larson RG (2005). A mean flow model for polymer and fiber<br />
turbulent drag reduction. App. Rheol. J., 15: 370-389.<br />
Shetty AM, Solomon MJ (2009). Aggregation in dilute solutions of high<br />
molar mass poly(ethylene) oxide and its effect on polymer turbulent<br />
drag reduction. Polym. J., 50: 261-270.<br />
Singh RD (1990). Encyclopedia of Fluid Dynamics, Vol. 9, Gulf<br />
Publishing Co., Houston, Texas, Chapt. 14: 425-480.<br />
Suali E, Abdul Bari HA, Hassan Z, Rahman MM (2010). The Study of<br />
Gycolic Acid Ethoxylate 4-nonylphenyl Ether on Drag Reduction. J.<br />
Appl. Sci., 10(21): 2683-2687.<br />
Toonder JMJD, Hulsen MA, Kuiken GDC, Nieuwstadt FTM (1997). Drag<br />
reduction by polymer additives in a turbulent pipe flow: Laboratory<br />
and numerical experiments. J. Fluid Mech., 337: 193-231.<br />
Virk PS, Merrill EW, Mickley HS, Smith KA (1967). The Toms<br />
phenomenon: Turbulent pipe flow of dilute polymer solutions. J. Fluid<br />
Mech., 30: 305-328.
Scientific Research and Essays Vol. 7(18), pp. 1777-1785, 16 May, 2012<br />
Available online at http://www.academicjournals.org/SRE<br />
DOI: 10.5897/SRE11.1759<br />
ISSN 1992-2248 © 2012 <strong>Academic</strong> <strong>Journals</strong><br />
Full Length Research Paper<br />
Available transfer capability and least square method<br />
Mojgan Hojabri 1 *, Mohammadsoroush Soheilirad 2 and Mahdi Hedayati 2<br />
1 Faculty of Electrical and Electronics Engineering, University Malaysia Pahang (UMP), 26600 Pekan, Pahang, Malaysia.<br />
2 Department of Electrical and Electronic Engineering, University Putra Malaysia (UPM), 43300Serdang, Selangor,<br />
Malaysia.<br />
Accepted 17 February, 2012<br />
In deregulated power industries, accurate and fast calculation of Available Transfer Capability (ATC) for<br />
seller in the electricity market is required. In this paper, deterministic ATC will be calculated using<br />
algebraic equation and linear optimization by Least Square (LSQR). Probabilistic ATC is also calculated<br />
by considering time varying load and load margin. The proposed method will be tested on IEEE 30 bus<br />
system. Deterministic ATC results will be compared with Benders, OPF and HCACO and probabilistic<br />
ATC results also will be compared with GA and HCACO.<br />
Key words: Available transfer capability (ATC), power system planning, LSQR.<br />
INTRODUCTION<br />
Competition in the electric power industry between sellers<br />
and buyers in power marketing poses new challenges for<br />
power system companies and researchers to find the<br />
best strategy for having beneficial energy trading. The<br />
technical challenges are related to the generation and<br />
transmission. Managing an effective operation can be<br />
provided by minimizing the operational cost, maximizing<br />
utilization of generators and transmission lines. However,<br />
power transmission systems are limited by the power<br />
transfer. The Available Transfer Capability (ATC) is<br />
required to be reported on the Open Access same–time<br />
Information System (OASIS) to inform all energy market<br />
participants of the maximum power transfer capability in<br />
power systems. Therefore to improve the power system<br />
efficiency and economy, a good strategy for calculating<br />
ATC is required. Moreover the strategy can be used to<br />
predict ATC for the future transmission enhancement in<br />
power system planning.<br />
Available transfer capability calculation is important for<br />
electric power companies and energy buyers. ATC<br />
calculation not only determines the energy transfer<br />
bounds but it also determines the reliability of the system<br />
in unsecured situations. Based on the definition of ATC<br />
by NERC in 1996, several researches have been done<br />
*Corresponding author. E-mail: mojgan.hojabri@gmail.com.<br />
on deterministic and probabilistic ATC calculations. Their<br />
objective was to find a fast and accurate method. From<br />
these researches, the speed of the deterministic ATC<br />
methods is better than probabilistic ATC calculation<br />
methods since in probabilistic ATC calculation<br />
uncertainties are considered.<br />
Probabilistic or stochastic power flow methods are used<br />
to accommodate the random nature of the operational<br />
load and generation data. Three important methods for<br />
stochastic power flow techniques are Monte Carlo,<br />
Convolution and Stochastic Algebraic. The Monte Carlo<br />
is a famous method to solve stochastic power flow<br />
problem (Huang and Yan, 2002; Yajing et al., 2005; Gao<br />
et al., 2006; Anselmo et al., 2007). This method by<br />
repeated trials of the deterministic ATC, calculates the<br />
probability distributions of the nodal powers, line flow and<br />
losses. The big disadvantage of Monte Carlo is<br />
computational burden.<br />
Convolution method calculated the impact of the<br />
uncertainty of load data to uncertainty of bus voltage and<br />
line power flow by (Borkowska, 1974). Some problem of<br />
this research is nonlinear relation between node leads<br />
and branch flows, and proper balance of generation and<br />
loads. In Stochastic power flow problem used (Dopazo et<br />
al., 1975), they assumed normally distributed generator<br />
for bus variables P and V. Then, they calculated power<br />
flows by using classical methods. Normality distributed<br />
complex random variables is the difficulty of this research
1778 Sci. Res. Essays<br />
as described in (Sauer, 1977; Feller, 1971).<br />
Stochastic Algebraic method (Stahlhut et al., 2005;<br />
Jonathan, 2007) is one of the earliest probabilistic ATC<br />
calculation methods. Using full AC load flow and linear<br />
algebraic equations make this method simple. To<br />
overcome the problem of using linear method for<br />
nonlinear problem, Krylov subspace method is used in<br />
this study as a power iterative mathematical method. In<br />
Stochastic Algebraic calculation, ATC is only calculated<br />
for bilateral transactions. The usefulness of a bilateral<br />
transaction is ATC can easily be calculated between two<br />
buses, where the transaction power enters and leaves<br />
the network without considering various margins like<br />
transmission reliability margin, capacity benefit margin<br />
etc. for ATC evaluation. Therefore ATC is approximately<br />
equal to Total Transfer Capability (TTC), which is a key<br />
component for ATC assessment. In this paper, ATC is<br />
determined for multilateral transaction based on linear<br />
optimization using LSQR method. The impact of other<br />
lines, generators and loads on power transfer could be<br />
taken into account. Then the ATC computation will be<br />
more realistic. Another benefit of this method is by using<br />
linear programming, which makes the ATC computations<br />
simple. Moreover, the nonlinear behavior of ATC<br />
equations are considered by using one of the best<br />
iteration methods called Krylov subspace method. It is a<br />
robust method that could handle the nonsymmetrical and<br />
definite program (Ioannis, 2007).<br />
METHODOLOGY<br />
ATC definition<br />
In this paper, ATC are defined by linear optimization. To maximize<br />
the ATC (Equation 1), the objective function for the calculation of<br />
ATC is formulated as (Gnanadass and Ajjarapu, 2008):<br />
Where� P and<br />
gi � P are total power generated in the<br />
gj<br />
sending and receiving area. And� Pli and� P are total<br />
lj<br />
power used in the sending and receiving area.<br />
The objective function measures the power exchange between the<br />
sending and receiving areas. The constraints involved include,<br />
a) Equality power balance constraint. Mathematically, each lossless<br />
bilateral transaction between the sending and receiving bus i must<br />
satisfy the power balance relationship.<br />
(1)<br />
(2)<br />
(3)<br />
For multilateral transactions, this equation is extended to:<br />
Where is the total number of transactions.<br />
b) Inequality constraints on real power generation and utilization of<br />
both the sending and receiving area.<br />
Where and are the values of the real power<br />
generation and utilization of load flow in the sending and receiving<br />
areas, and are the maximum of real power<br />
generation and utilization in the sending and receiving areas.<br />
c) Inequality constraints on power rating and voltage limitations.<br />
With use of algebraic equations based load flow, margins for ATC<br />
calculation from bus i to bus j are represented in Equations (7and 8)<br />
and Equations (10 and 11). For thermal limitations the equations<br />
are,<br />
Where is determined as in Equation 8.<br />
Where and are bus voltage of the sending and receiving<br />
areas. And is the reactance between bus i and bus j. For<br />
voltage limitations,<br />
(4)<br />
(5)<br />
(6)<br />
(7)<br />
(8)<br />
(9)<br />
(10)<br />
(11)<br />
Where and are calculated as (Jonathan,<br />
2007):<br />
(12)
(13)<br />
Where is susceptance and represents a<br />
diagonal matrix whose elements are (for each transmission<br />
line), L is the incident matrix, PF is the power factor, and E11, E12,<br />
E21 and E22 are the sub matrixes of inverse Jacobian matrix. This<br />
can be achieved by steps descrided later. Reactive power (Q)<br />
constraints must be considered as active power constraints in<br />
Equations 3 to 6.<br />
Due to nonlinear behavior of power systems, linear<br />
approximation and can yield<br />
errors in the value of the ATC. In order to get a more precise ATC,<br />
an efficient iterative approach must be used. One of the most<br />
powerful tools for solving large and sparse systems of linear<br />
algebraic equations is a class of iterative methods called Krylov<br />
subspace methods. The significant advantages are low memory<br />
requirements and good approximation properties. To determine the<br />
ATC value for multilateral transactions the sum of ATC in Equation<br />
14 must be considered,<br />
Where k is the total number of transactions.<br />
(14)<br />
Krylov subspace methods form the most important class of iterative<br />
solution method. Approximation for the iterative solution of the<br />
linear problem for large, sparse and nonsymmetrical Amatrices,<br />
started more than 30 years ago (Adam, 1996). The<br />
approach was to minimize the residual r in the<br />
formulation . This led to techniques like,<br />
Biconjugate Gradients (BiCG), Biconjugate Gradients Stabilized<br />
(BICBSTAB), Conjugate Gradients Squared (CGS), Generalized<br />
Minimal Residual (GMRES), Least Square (LSQR), Minimal<br />
Residual (MINRES), Quasi-Minimal Residual (QMR) and Symmetric<br />
LQ (SYMMLQ).<br />
The solution strategy will depend on the nature of the problem to<br />
be solved which can be best characterized by the spectrum (the<br />
totality of the eigenvalues) of the system matrix A. The best and<br />
fastest convergence is obtained, in descending order, for A being:<br />
(a) Symmetrical (all eigenvalues are real) and definite,<br />
(b) Symmetric indefinite,<br />
(c) Nonsymmetrical (complex eigenvalues may exist in conjugate<br />
pairs) and definite real, and<br />
(d) Nonsymmetrical general<br />
However MINRES, CG and SYMMLQ can solve symmetrical and<br />
indefinite linear system whereas BICGSTAB, LSQR, QMR and<br />
GMRES are more suitable to handle nonsymmetrical and definite<br />
linear problems (Ioannis, 2007). The ATC margins equations can<br />
be represented in the general form:<br />
(15)<br />
Where represents vector form (number of branches) from<br />
Equations 7 and 8 and also vector form (number of buses)<br />
of Equations (10 and 11). With iteration step k, Equation 15 gives<br />
the residual r k.<br />
And the linearized form is:<br />
Hojabri et al. 1779<br />
(16)<br />
(17)<br />
Where A represents or<br />
in diagonal matrix form (number of branches)<br />
x (number of branches) or (number of buses) x (number of buses),<br />
and b gives or in vector<br />
form (number of branches) and or<br />
in vector form (number of buses) while the inequalities (7, 8, 10 and<br />
11) can be rewritten as in Equations 18-21. In this case, the nature<br />
of A is nonsymmetrical and definite. However, all of the Krylov<br />
subspace methods can be used for ATC computation but<br />
BICGSTAB, LSQR, QMR and GMRES are more suitable to handle<br />
this case.<br />
(18)<br />
(19)<br />
(20)<br />
(21)<br />
In this paper, LSQR is used for ATC computation. Numerically,<br />
LSQR is more reliable in various circumstances than the other<br />
Krylov subspace methods (Christopher and Michael, 1982). Small<br />
residual and using the standard QR factorization are other<br />
advantages of LSQR method (Golub and Kahan, 1965).<br />
Statistical analysis<br />
Statistical moments are used to provide some sort of measure for a<br />
probability distribution of ATC. The most important and useful<br />
moment is the center of a distribution of X. This center is called the<br />
mean, and is usually denoted as the , or the expectation of<br />
random variable ATC (Daniel and Ralph, 1996),<br />
(22)<br />
The mean is just one measure that a probability distribution has.<br />
The variance is another popular statistical measure of a probability<br />
distribution. The variance is a measure of the spread of the<br />
distribution. The variance is typically symbolized as<br />
or . The square root of variance is called standard
1780 Sci. Res. Essays<br />
deviation .<br />
Figure 1. IEEE 30 Bus system.<br />
Table 1. Basic statistics for expected ATC (IEEE 30 bus system).<br />
Indices<br />
T1<br />
Line Outage<br />
T2 T3<br />
Time Varying Load<br />
T1 T2 T3<br />
Mean 103.6 96.76 44.437 106.81 102.93 48.034<br />
Standard Deviation 8.36 12.72 5.592 2.61 2.41 0.579<br />
Skewness -1.9 -2.6 -1.84 0.06 0.06 0.07<br />
Kurtosis 4.06 8.82 4.3 -1.53 -1.53 -1.52<br />
(23)<br />
(24)<br />
Where is the mean of ATC and N is the number of data. Two<br />
additional measures are used with the mean and standard deviation<br />
to help describe a probability distribution. These additional<br />
measures are skewness and kurtosis. The skewness is defined as:<br />
(25)<br />
When is the third moment of the mean of ATC and is the<br />
standard deviation. The measure of skewness is often useful for<br />
nonsymmetrical distribution. If the tail of the distribution is longer on<br />
the right, the skewness is a positive number. The skewness of a<br />
normal distribution is zero. The kurtosis of ATC is defined as in<br />
Equation 26. The kurtosis is a measure of peakedness of a<br />
distribution. A distribution that has a high kurtosis can range from -2<br />
to . The kurtosis of a normal distribution is 3 (Joanes and Gill,<br />
1998).<br />
When is the fourth moment of the mean.<br />
RESULTS<br />
(26)<br />
Figure 1 shows the IEEE 30 bus system which is<br />
separated into three areas. Power must be transferred<br />
among these areas by three interconnection paths.<br />
Based on Figure 1, these transaction paths which are<br />
called T1 (between area 1 and area 2), T2 (between area<br />
1 and area 3) and T3 (between area 2 and area 3)<br />
contain several lines. These lines are connected between<br />
sender buses and receiver buses as shown in Figure 1.<br />
Power could be transferred between sender and receiver<br />
areas by these transfer lines. Multilateral probabilistic<br />
ATC calculation is done for this system and the statistical<br />
analysis is described for 3 different areas for the IEEE 30<br />
bus system.<br />
Statistical analysis was done for ATC based on line<br />
outage and time varying load. The results are shown in<br />
Table 1. According to these results, the ATC mean and<br />
ATC standard deviation comparison calculated by these
Figure 2. Mean comparison for ATC based on line outage and time varying load.<br />
Figure 3. Standard deviation comparison for ATC based on line outage and time<br />
varying load.<br />
two methods for IEEE 30 bus system. Figures 2 and 3<br />
indicate that the mean of ATC based on time varying load<br />
is more than the ATC mean based on line outage for all<br />
transaction paths and the standard deviation for time<br />
varying load is also less than line outage. In overall, the<br />
probabilistic ATC based on time varying load is more<br />
reliable than probabilistic ATC based on line outage. It is<br />
related to having bigger mean and smaller standard<br />
deviation. Therefore, it can be concluded that more<br />
power transferred with more reliability can be contracted<br />
Hojabri et al. 1781<br />
between seller and buyer of energy when ATC is<br />
estimated based on time varying load.<br />
Based on the obtained histogram (Figure 4), it can be<br />
concluded when line outage occurred, ATC intends to go<br />
right side of curves (skewness is negative). These<br />
histograms indicate the ATC do not follow normal curve.<br />
However in time varying load (Figure 5), the ATC is<br />
distributed around the mean (skewness is close to zero)<br />
and their histograms show the ATC follows1 the normal<br />
curve.
1782 Sci. Res. Essays<br />
DISCUSSION<br />
(a)<br />
(b)<br />
(c)<br />
Figure 4. ATC Histogram for Line Outage; (a) Area1 – Area2 (T1), (b) Area1 -<br />
Area3 (T2), (c) Area2 - Area3 (T3).<br />
For IEEE 30 bus system, ATC calculations are compared<br />
against Optimal Power Flow (OPF) (Xiong and Guoyu,<br />
2005), Benders (Shaaban et al., 2003; Weixing and<br />
Xiaoming, 2008) and Hybrid Continuous Ant Colony<br />
Optimization (HCACO) (Guoqing et al., 2008). These are<br />
optimization methods which considered most of the<br />
limitations. The percent difference (Diff %) are<br />
determined for Krylov Algebraic Method (KAM) result<br />
against each of these methods.<br />
(27)<br />
The percentage differences are shown in columns 3, 5<br />
and 7 of Table 2. Comparing KAM to Benders and OPF,
(a)<br />
(b)<br />
Figure 5. ATC Histogram for Time Varying Load; (a) Area1 – Area2 (T1), (b) Area1 -<br />
Area3 (T2), (c) Area2 - Area3 (T3).<br />
the amount of ATC has improved except for T2 by<br />
Benders. Comparing with HCACO, KAM results are less<br />
than HCACO with 7.78, 0.47 and 6.12 percent<br />
differences for T1, T2 and T3. The small percent<br />
(c)<br />
Hojabri et al. 1783<br />
difference proves that the deterministic ATC results of<br />
KAM compared well to these methods. Since the amount<br />
of ATC for T3 calculated by Benders is very small the<br />
difference between KAM and Benders in this case is big.
1784 Sci. Res. Essays<br />
Table 2. Deterministic ATC comparison results for IEEE 30 Bus system.<br />
Transaction Paths<br />
Benders Diff% OPF<br />
ATC (MW)<br />
Diff% HCACO Diff% KAM<br />
T1 104.19 2.48 101.5 3.45 115.46 7.78 106.81<br />
T2 103.31 0.35 96.96 4.04 103.43 0.47 102.95<br />
T3 32.21 39.43 47.59 0.61 45.18 6.12 48.03<br />
Table 3. ATC Statistical comparison for IEEE 30 bus systems.<br />
Sending area to<br />
receiving area<br />
GA HCACO KAM (Proposed method)<br />
Mean Standard<br />
deviation<br />
Mean Standard<br />
deviation<br />
Mean Standard<br />
deviation<br />
T1 106.54 2.36 115.46 2.42 106.81 2.61<br />
T2 98.85 3.50 103.43 2.33 102.93 2.41<br />
T3 42.82 1.49 45.18 0.45 48.03 0.58<br />
However, this result is far from the OPF and HCACO with<br />
38.55 and 33.52% differences.<br />
The mean and standard deviation of ATC based on<br />
KAM (Proposed method), GA (Genetic Algorithm) and<br />
HCACO are shown in Table 3 for IEEE 30 bus system.<br />
Columns 2, 4 and 6 of this table show the ATC mean of<br />
T1, T2 and T3 transactions. The related standard<br />
deviations are also shown in columns 3, 5 and 7 of this<br />
table. Based on this Table, the mean of ATC for KAM is<br />
bigger than GA. However its standard deviation is less<br />
than GA except for T1. In overall, the result of KAM is<br />
better and more reliable than GA. Compared to HCACO,<br />
the mean and standard deviations of KAM are smaller<br />
and bigger.<br />
Conclusion<br />
ATC calculation can be divided into two categories,<br />
deterministic ATC and probabilistic ATC. In this paper<br />
ATC was calculated for IEEE 30 bus systems which took<br />
into account the full AC load flow, linear optimization,<br />
thermal and voltage constraints. The results of ATC<br />
computation obtained from MATLAB programming were<br />
used to estimate the ATC by using Minitab software by<br />
considering time varying load and line outage for power<br />
system planning.<br />
To verify the proposed method, the results of<br />
deterministic and probabilistic ATC for IEEE 30 bus<br />
system were compared to other methods. The<br />
deterministic result with small percent difference is close<br />
to HCACO. However the amount of deterministic ATC for<br />
KAM is bigger than Benders and OPF. Lower standard<br />
deviation and higher mean of ATC based on time varying<br />
load indicate that the time varying load is more suitable<br />
than line outage to identify the probabilistic ATC. The<br />
main statistical results for IEEE 30 bus system were also<br />
compared with GA and HCACO methods. According to<br />
the results, the ATC standard deviation of KAM is less<br />
than the GA algorithm and this result is close to HCACO<br />
which was determined based on random search<br />
technique.<br />
REFERENCES<br />
Adam S (1996). Fundamental Concepts of a Krylov Subspace Power<br />
Flow Methodology. IEEE Trans. Power Syst., 11(3): 1528-1537.<br />
Anselmo B, Rodrigues G, Silva MD (2007). Probabilistic Assessment of<br />
Available Transfer Capability Based on Monte Carlo Method with<br />
Sequential Simulation. IEEE Trans. Power Syst., 22(1): 484-492.<br />
Borkowska B (1974). Probabilistic Load Flow. IEEE Trans. Power Appl.<br />
Syst., PAS93(4): 752-759.<br />
Christopher C, Michael A (1982). LSQR: An Algorithm for Sparse Linear<br />
Equations and Sparse Least Squares. ACM Trans. Math. Softw.,<br />
8(1): 43-71.<br />
Daniel S, Ralph BD (1996). Practical Engineering Statistics. Canada:<br />
John Wiley and Sons.<br />
Dopazo J, Klitin O, Sasson A (1975). Stochastic load flows. IEEE Trans.<br />
Power Syst., PAS-94(2): 299- 309.<br />
Feller W (1971). An Introduction to Probability Theory and Its<br />
Applications, Second Edition. New York: John Wiley and Sons.<br />
Gao Zhou, YM, Li G (2006). Sequential Monte Carlo Simulation Based<br />
Available Transfer Capability Calculation. Int. Conf. Power Syst.<br />
Technol., pp. 1-6.<br />
Gnanadass R, Ajjarapu V (2008). Assessment of Dynamic Available<br />
Transfer Capability using FDR PSO Algorithm. Elektrika J. Elect.<br />
Eng., 10(1): 20-25.<br />
Golub G, Kahan W (1965). Calculating the Singular Values and<br />
Pseudoinverse of A Matrix. SIAM J. Numer. Anal., 2: 205-224.<br />
Guoqing L, Zhiyuan L, Hao S (2008). Study of Available Transfer<br />
Capability Based on Hybrid Continuous Ant Colony Optimization .<br />
IEEE Conf. Elect. Utility Deregul. Restruct. Power Technol., pp. 984-<br />
989.<br />
Huang GM, Yan P (2002). Composite System Adequacy Evaluation<br />
Using Sequential Monte Carlo Simulation for Deregulated Power<br />
Systems.IEEE Power Engineering Society Summer Meeting.<br />
Ioannis K A (2007). Computational Theory of Iterative Methods (Vol.<br />
15).Elsevier.
Joanes DN, Gill CA (1998). Comparing Measures of Sample Skewness<br />
and Kurtosis. J. Royal Stat. Soc. (Series D): The Stat., 47(1): 183-<br />
189.<br />
Jonathan WS (2007). Stochastic-Algebraic Calculation of Available<br />
Transfer Capability. IEEE Trans. Power Syst., 22(2): 616-623.<br />
Sauer P (1977). A Generalized Stochastic Algorithm.W.Lafayette: Ph.D<br />
Thesis, Purdue University.<br />
Shaaban M, Li WX, Yan H, Ni YX, Wu FF (2003). ATC Calculation with<br />
Steady-State Security Constraints using Benders Decomposition.<br />
IEEE Proc. Generat. Transm. Distr., pp. 611-615.<br />
Stahlhut JW, Heydt GT, Sheblé GB (2005). A Stochastic Evaluation of<br />
Available Transfer Capability. IEEE PES General Meeting.<br />
Weixing AL, Xiaoming M (2008). A Comprehensive Approach for<br />
Transfer Capability Calculation Using Benders Decomposition.<br />
DPRT, pp. 301-306.<br />
Hojabri et al. 1785<br />
Xiong P, Guoyu (2005). Available Transfer Capability Calculation<br />
Considering Voltage Stability Margin. Elect. Power Syst. Res., 76(1-<br />
3): 52-57.<br />
Yajing G, Ming Z, Gengyin L, Yuekui H, Limin X, Rui L (2005). Monte<br />
Carlo Simulation Based Available Transmission Capability<br />
Calculation. IEEE/PES Transm. Distr. Conf. Exhibit. Asia Pac., pp.1-<br />
6.
Scientific Research and Essays Vol. 7(18), pp. 1786-1797, 16 May, 2012<br />
Available online at http://www.academicjournals.org/SRE<br />
DOI: 10.5897/SRE12.097<br />
ISSN 1992-2248 © 2012 <strong>Academic</strong> <strong>Journals</strong><br />
Full Length Research Paper<br />
Seasonality of non-timber forest products in the Kupe<br />
mountain region of South Cameroon<br />
Ngane B. K.*, Ngane E. B., Sumbele S. A., Njukeng J. N., Ngone M. A. and Ehabe E. E.<br />
Institute of Agricultural Research for Development (IRAD), Ekona Regional Research Centre, Buea, South West Region,<br />
Cameroon.<br />
Accepted 23 April, 2012<br />
Non-timber forest products (NTFPs) are economically-important commodities, especially in the humid<br />
and semi-humid regions of many sub-Saharan countries. However, in many of these countries, there is<br />
little information on the seasonality characterising their availability and collection, rendering difficult<br />
the development of meaningful packages for their sustainable use and valorisation. A study was<br />
therefore conducted in randomly selected villages in three zones around the sub-montane forests of<br />
Mount Kupe in South West Cameroon, in order to evaluate the effects of seasons on the collection and<br />
availability of these products. Data were collected through administration of semi-structured<br />
questionnaires and a rapid rural appraisal. Most NTFPs (85%) showed seasonality to different extents,<br />
in availability and abundance. Although, considerable differences were observed in the proportion of<br />
respondents who collected NTFPs during the dry and rainy seasons, no such difference could be<br />
attributed to the climatic or geographical zones. Half of the respondents collected NTFPs during the<br />
rainy season when there was relatively less work in their farms. About one-sixth (15.2%) collected<br />
NTFPs during the dry season (peak agricultural period) and a third (35%) collected some NTFPs all year<br />
round.<br />
Key words: Cameroon, humid forest, non-timber, seasonality, collection.<br />
Introduction<br />
Until recently, timber products like industrial and derived<br />
sawn timber, wood chips, wood-based panel and pulp<br />
have been considered as the major or even the most<br />
important forest products due to their income-generating<br />
capacity. Meanwhile, other forest products, especially<br />
those used as food, medicinal plants, household and<br />
agricultural implements, etc. have traditionally been considered<br />
of minor or secondary importance (Makon et al.,<br />
2005). The importance of these later materials, extracted<br />
from forest ecosystems, has increased considerable<br />
today following their use within households and as they<br />
are equally traded for their social, cultural or religious<br />
significance (Ndoye and Tieguhong, 2004; Ngwasiri et<br />
al., 2005). Unlike some authors who term these materials<br />
*Corresponding author. E-mail: nganekoben@yahoo.co.uk. Tel:<br />
(+237)75936380.<br />
as non-wood forest products (FAO, 1991; Sunderland<br />
and Oboma, 1999; Vantomme, 1999), the term nontimber<br />
forest products (NTFPs) used by other authors<br />
(Falconer, 1992; Kio and Abu, 1993; Okafor, 1989) is<br />
preferred in this study since some of the products such<br />
as axe handles, fuel wood and chewing sticks contain<br />
lignin (wood).<br />
Most NTFPs are not domesticated and are just<br />
collected from the wild. They are often not available yearround<br />
or when available, are in limited quantity. The<br />
exploitation of NTFPs could therefore be regarded as a<br />
seasonal activity that contributes to improving the<br />
livelihoods of rural populations where prices of major<br />
agricultural products are low, fluctuating and seasonal.<br />
Agriculture is the mainstay of the people of the Kupe<br />
area. Farming is carried out mainly for subsistence and<br />
the traditionall farming system is shiefting cultivation<br />
which is wasteful to land. Arable crops, tree crops and<br />
animals are being farmed. Major arable crops are
Table 1. Market value of some important NTFPs in Southern Cameroon.<br />
Species name<br />
% of harvest<br />
sold<br />
Total revenue<br />
(CFA) in 4 months<br />
Total revenue ($ US)<br />
in 4 months<br />
Coula edulis 5 15000 3.1<br />
Cola acuminate/nitida 30 2500 5.1<br />
Cola pedidota 30 3800 7.8<br />
Dacryodes edulis 10 11000 22.5<br />
Dacryodes macrophylla 10 2000 4.1<br />
Elaieis guinensis ‹1 2800 5.7<br />
Garcinia lucida 10 400 0.8<br />
Irvingia<br />
wombolu<br />
gabonensis/<br />
20 79000 161.6<br />
Ricinodendron heudelotti 100 5800 11.9<br />
Strophanthus gratis 100 32500 66.5<br />
cocoyam, plantains, banana, cassava, yam, maize,<br />
cabbage, pepper and okro. Tree crop agriculture involves<br />
the growing of cacao, coffee and oil palm. Animal farming<br />
includes the rearing of goats, sheep and birds. Fishing of<br />
tadpoles and frogs is carried out in in-land water bodies.<br />
Disturbing and persistent drop in the prices of cocoa<br />
and coffee (major export crops of the area) for two<br />
decades now have further worsened poverty levels in this<br />
area. However the sale of food crops now constitutes a<br />
major income source particularly as foreign buyers from<br />
mostly Gabon and Equitorial Guinea have flooded the<br />
market. NTFPs usually collected in the study area are<br />
used for food, medicine, household utensils and<br />
construction material.<br />
Due to slightly different vegetation and climatic<br />
conditions in the three zones considered in the study, it is<br />
expected that species variation and variation in time of<br />
production of NTFPs will be observed (Leakey et al.,<br />
2000). For example, off season Dacryodes edulis could<br />
be found in Kupe East. Also Gnetum spp. (Figure 4L)<br />
which is of local, national, regional and international<br />
significance could be found only in Kupe East with higher<br />
temperatures throughout the year. With local variation in<br />
the vegetation and climatic conditions in the study area, it<br />
is possible that a particular product could be obtained<br />
from different species in different zones.<br />
Although most of the NTFPs are collected for<br />
household consumption, some are traded to suppliment<br />
the cash income of households. While NTFPS such as<br />
Ricinodendron heudelotti and D. edulis are traded locally<br />
in village markets, others such as Irvingia spp. and<br />
Gnetum spp. are taken to city markets and are bought by<br />
buyers who export to Nigeria, Europe and the United<br />
States of America.<br />
In the Kupe area, collection of NTFPs is an income<br />
diversification activity which is risk-reducing and at the<br />
same times a way of coping with constraints in the main<br />
income generating activity of the area, agriculture.<br />
Important NTFPs such as Garcinia cola (Figure 4a), Piper<br />
Ngane et al. 1787<br />
guinense (Figure 4E, F and I) and R. heudelotti that play<br />
a major role in generating income for the local people are<br />
not available throughout the year. The scarcity of such<br />
NTFPs during off-seasons leads to food insecurity and<br />
economic vulnerability. This adversely affects the socioeconomic<br />
life of the community as malnutrition and<br />
associated diseases coupled with poverty reduce the<br />
living standard of the people.<br />
The absence of a particular medicinal NTFP such as<br />
Afromonium melegueta (Figure 4J) at a certain period of<br />
the year can also result to serious health problems in the<br />
communities that depend on it for the cure of various<br />
ailments. Dijks Van (1999) showed that NTFPs collection<br />
and production can provide a source of livelihood as<br />
shown on Table 1.<br />
In such an area where NTFPs play a major role in<br />
enhancing status and wellbeing of the local people,<br />
where there is ethnic diversity, local variation in climatic<br />
conditions with consequent variation in use and<br />
availability of NTFPs during different seasons of the year,<br />
there is need for relevant, reliable and timely information<br />
on their seasonal availability and estimates of their stocks<br />
if adequate forecasts of use, marketing and sustainability<br />
are to be made.<br />
The objective of this study was therefore to provide<br />
information on the true value and significance of nontimber<br />
forest products during different seasons of the<br />
year in the Mount Kupe area. More specifically this study<br />
attempts to:<br />
i. Identify the various types of non-timber forest products<br />
in the study area during the different seasons.<br />
ii. Assess the factors affecting the involvement of local<br />
populations in their collection.<br />
iii. Proffer recommendations on management strategies<br />
that will mitigate the effect of seasonality of these<br />
products.<br />
Based on the objectives of this study the hypothesis
1788 Sci. Res. Essays<br />
tested were as follows:<br />
i. That the availability of NTFPs depends on different<br />
seasons of the year.<br />
ii. That the availability of NTFPs depends on zones.<br />
iii. That NTFPs contribute to the socio-economy of the<br />
local people during different seasons of the year.<br />
MATERIALS AND METHODS<br />
Survey locus<br />
The Kupe area straddles two regions of Cameroon, the South West<br />
and Littoral regions. The population that inhabits the Kupe area<br />
represents an array of socio-cultural, economic and political<br />
identities possessing a diverse ethnic originality made up of various<br />
indigenous Bantu clans (Muetuck, Mbo, Manehas, Muasundem<br />
etc), settlers from the North West Region and immigrant Bamilikis<br />
from the West Region of Cameroon. This people make up a<br />
population of about 100,000.<br />
Forests that make up the Mount Kupe are mostly sub-montane<br />
equatorial rain forest and are known to support a wide variety and<br />
abundance of resources which are of huge importance to the<br />
people that inhabit the area for livelihood improvement, recreation,<br />
cultural promotion, etc. Forests resources present include timber,<br />
NTFPs, ethno-botanical resources, etc. The area is also known to<br />
have huge eco-tourism potentials as well as very rich cultural<br />
values. Moreover the quality of the forest provides for undeniable<br />
environmental services including water catchments.<br />
The survey was conducted through out the year 2008 in the<br />
Mount Kupe area located within the humid forests of Southern<br />
Cameroon. The Kupe Mountain spans the borders of the South<br />
West and Littoral Regions of Cameroon. The area, home to the<br />
Bakossi tribe and a high proportion of non-indigenes, could be<br />
partitioned into three zones populated by three different clans with<br />
distinctively different eating habits: Manehas in the East<br />
(Francophone), Muetuck in the West (Anglophone) and<br />
Muasundem in the North- Anglophone (Ejedepang-Koge, 1986).<br />
These zones are different with respect to their altitudes,<br />
geographical positions, climate and vegetation.<br />
The difference in altitudes and geographical position has led to<br />
differences in the climate as well as the vegetation of the different<br />
zones. The climate of this area is of the equatorial or transitional<br />
tropical type. Despite some local variations, there is a zonal<br />
uniformity of high and constant temperatures. Rainfall varies from<br />
2500 mm per annum in Kupe East, to about 5000 mm per annum in<br />
the entire area. Minimum monthly temperatures vary from 19 to<br />
21°C; though much lower in Kupe West (2064 m) and maximum<br />
from 30.55 to 35.20°C. Relative humidity is about 85%.<br />
Generally two main seasons could be distinguished separated by<br />
an intermediate season. (i) Mid-June to October: a major rainy<br />
season of overcast and misty weather, continuous rains and<br />
drizzling; (ii) Mid-November to Mid March: a major dry season of dry<br />
foggy weather, harmattan and diffuse radiance, (iii) Mid October to<br />
mid-November/mid-March to mid June: intermediate/transition<br />
season of changeable weather, storms and rain showers alternating<br />
with bright intervals.<br />
Sampling of respondents and data collection<br />
The studied area was stratified based on the zones around the<br />
Kupe mountain (East, West and North of the mountain). Three<br />
clans were randomly selected from the entire Kupe area, and three<br />
villages randomly selected from each clan, as shown as follows:<br />
i. Kupe East zone (Manehas clan): Nlohe, Lala and Kola villages<br />
(French speaking villages);<br />
ii. Kupe West zone (Muetuck Clan): Nyasoso, Ngusi and Atop<br />
villages (English speaking villages);<br />
iii. Kupe West zone (Muasumdem clan): Mpako, Tape-Etube and<br />
Bendume villages (English speaking villages).<br />
A semi structured questionnaire was used as the main survey<br />
instrument for information collection. It was designed in five parts A<br />
to E. Section A was designed to elicit information on the<br />
background of the villages. Section B dealt with the biodata of the<br />
respondents. Section C sought to elicit information on the<br />
identification of NTFPs. Section D was to gather information on<br />
seasonal variation in the collection of NTFPs and section E dueled<br />
on the socioeconomic contribution of NTFPs.<br />
Before the actual administration of the questionnaire a<br />
reconnaisance survey was undertaken during which some<br />
background information on the study area was collected. A pre-test<br />
survey was also undertaken. During this exercise test questionnaire<br />
were randomly administered in three villages which though different<br />
from the villages selected for the actual study were within the same<br />
locality. The <strong>complete</strong>d test questionnaire were collated and<br />
analysed. Adjustments were made before the final production of<br />
questionnaire before the survey proper was undertaken.<br />
Ten percent of the estimated number of households in the<br />
selected villages was calculated (Hannagan, 1989). 280<br />
questionnaire were filled and 250 returned as follows: 66, 87, and<br />
97 for Kupe East, Kupe North and Kupe West respectively. In<br />
addition to this, a rapid rural appraisal (RRA) was carried out - using<br />
transect walks, structured direct observations and key informant<br />
interviews.<br />
Data analyses<br />
The data were statistically analysed for the sources of variance on<br />
a randomised <strong>complete</strong> block design using the M-stat software<br />
(Alika, 1997). This was complimented by descriptive statistics using<br />
the Sphinx package.<br />
Results<br />
This survey showed that non-timber forest products were<br />
obtained from about sixty plant species and twenty<br />
animal genera (Tables 2 to 4). The majority of<br />
respondents collected NTFPs during the rainy season<br />
(Figure 2). On the average for the three zones, about half<br />
(50.4%) of the respondents collected NTFPs during the<br />
rainy season, while 15.2% collected NTFPs during the<br />
dry season. Also, 34.4% collected some NTFPs all the<br />
year round. Significant differences (P < 0.05) were<br />
observed between the number of respondents collecting<br />
NTFPs during the rainy and dry seasons but there was<br />
no significant difference between those collecting in the<br />
rainy season and those collecting all year round. There<br />
was a significant difference in the number of respondents<br />
collecting NTFPs during the different seasons in the<br />
different zones of the study area. The main reason for<br />
much collection of NTFPs during the rainy season was as<br />
a result of less farm work during this period (Table 5);<br />
though most of the NTFPs show seasonality in their<br />
collection (Tables 2, 3 and 4). Some human food NTFPs
show stark variation in collection during the different<br />
seasons of the year (Table 2). For example, though<br />
bushmeat (game) is generally harvested more frequently<br />
during the rainy season, pestilential species were caught<br />
relatively more frequently during the dry season. In the<br />
Eastern side of Kupe, fruit production was much later<br />
(end of season) and off-season fruit were available for<br />
collection. Construction materials show much less<br />
seasonality. However most construction and maintenance<br />
of houses, fences and huts takes place during<br />
the dry season (Table 3). Non-timber forest products also<br />
provide a source of livelihoods for rural people. This<br />
employment is basically on part-time basis (Figure 3).<br />
The situation was similar in the different zones but more<br />
full-time employment was provided by NTFP activities<br />
during the rainy season than during the dry season.<br />
Discussion<br />
The reason for collection of NTFPs in the different<br />
seasons was tested and revealed that most respondents<br />
collect NTFPs during the rainy season because of less<br />
work in their farms. This was also observed by<br />
Sunderland and Oboma (1999).<br />
This can further be attributed to the fact that during the<br />
dry season most labour is devoted to agricultural<br />
activities, while during agricultural slack periods much<br />
time is allocated to NTFP collection. This sometimes<br />
leads to over harvesting and unsustainable harvesting of<br />
NTFPs. Biliso and Lojoly (2006) also posited that there<br />
are no other significant sources of income during the<br />
rainy season. During the peak of the rainy season<br />
(hungry period), stored food supplies dwindle, most food<br />
crops on the farm are yet to mature and as such there is<br />
scarcity of food. NTFPs are important during this period<br />
as substitutes for staple foods, accompaniments and<br />
snacks to supplement the food crop supplies. When not<br />
available the local people are deprived of important food<br />
nutrients, medicines and household income. This has a<br />
very serious adverse effect in Kupe West and Kupe North<br />
than in Kupe East where off-season products are<br />
available and commercialisation of NTFPs more lucrative<br />
due to access to major city markets.<br />
Some respondents showed that harvesting periods of<br />
most NTFPs in Kupe coincide with school resumption<br />
(September in rainy season), hence NTFPs are highly<br />
needed to raise income for fees, uniforms and purchase<br />
of books. Buyungu (2000) also observed that this was<br />
also the situation in other West African countries like<br />
Ghana, Nigeria and Cote d’Ivoire. Some NTFPs such as<br />
D. edulis and forest vegetables are very much perishable<br />
hence more seasonal than bush mango (Irvingia<br />
gabonensis and wombulu Figure 4D) which can be dried<br />
to be used when not in season. This has also been<br />
observed by Leakey et al. (2000). It was also observed<br />
that Irvingia gabonensis was available in the rainy season<br />
Ngane et al. 1789<br />
(June-September) while Irvingoa wombulu was available<br />
during the dry season (October to February) making the<br />
product, ogbono available all year round. It was also<br />
observed that off-season fruits could be found in the<br />
Eastern side of Mount Kupe probably due to dryer<br />
climatic conditions in the leeward side of the mountain.<br />
Another important reason for much NTFPs collection<br />
during the rainy season is that very important NTFPs<br />
such as D. edulis (plum or safou), G. cola (bitter kola,<br />
Figure 4G) and Cola acuminate and nitida (kolanuts,<br />
Figure 4A) mature during the rainy season.<br />
The study showed that in the heart of the dry season,<br />
the availability of snails dwindles as they go into<br />
aestivation and are sometimes destroyed by wild fires.<br />
More forest vegetables were collected during the dry<br />
season as a result of scarcity of vegetables from farms.<br />
According to Hart et al. (2005), the harvesting of forest<br />
vegetables is more during the dry season because these<br />
serve as alternatives to cultivated vegetables which are<br />
perishable and scarce at this period. Though more game<br />
was obtained during the rainy season, pestilential species<br />
of bushmeat are caught during the dry season when their<br />
habitats are destroyed during land preparation for the<br />
next planting season. In this study, it was observed that<br />
the availability of some forest spices such as<br />
Ricinodendron heluidothii (njangsang) and Tetrapleura<br />
tetraptera (four corner or essisang) show very little<br />
seasonal variation in availability because they generally<br />
keep well, if dried. Value is added when preserved and<br />
are sometimes exported to neighbouring Nigeria. This<br />
contributes immensely to improving the living standard of<br />
the local people.<br />
Palm wine consumption largely depended on seasonal<br />
availability and perishability and is mostly drunk during<br />
the dry season when tapping is easy, without dilution<br />
from rain water. Local gin (locally called “afofo” and<br />
“odontol”) mostly produced from palm wine is drunk more<br />
during the rainy season when palm wine is relatively<br />
scarce. People tend to drink more of this local gin during<br />
the rainy season to have themselves warmed up.<br />
The collection of non-timber forest products used as<br />
medicine, construction materials and household items<br />
was much less seasonal because the parts mostly used<br />
for these purposes are the bark (cortex) and wood which<br />
are less perishable and are available throughout the year.<br />
Also in the case of some species like bitter cola (G. cola)<br />
and ‘ngou’ (Garcinia lucida) whose fruits are used as<br />
medicine, the bark is taken as a good substitute for<br />
treating the same ailments when the fruits are not in<br />
season. This also has a destructive effect on the tree<br />
which is sometimes girdled to death, especially as this<br />
occurs even in the dry season. Much harvesting of<br />
medicinal plants was observed in Kupe North because<br />
Croton spp. (“Ndume”, Figure 4B) which is considered by<br />
the Bakossi ethnic group as the most important medicinal<br />
plant is abundant in Bendume village from which the<br />
village derived her name (Bendume meaning plenty of
1790 Sci. Res. Essays<br />
Table 2. Seasonality of major food NTFPs in the Kupe area.<br />
Source Product<br />
Usefulness in the study area<br />
Rai<br />
ny<br />
Kupe West Kupe North Kupe East<br />
Dry<br />
All<br />
year<br />
Rainy Dry All year Rainy Dry All year<br />
(a) – Food of plant<br />
origin<br />
Elaeis guineensis<br />
Palm oil and kernel Cooking oil 2 3 2 1 2 1 2 3 2<br />
Palm wine Alcoholic drink 1 3 1 1 3 1 1 1 1<br />
Raphia hookery Palm wine Alcoholic drink 2 3 2 1 2 1 3 1 1<br />
Spices and condiments<br />
Ricinodendron heudelotii Njasang Condiment, vitamin c 3 1 1 3 1 1 2 1 1<br />
Irvingia gabonensis Ogbono condiment 2 0 0 2 0 0 1 0 0<br />
Irvingia wombulu Ogbono condiment 0 2 1 0 1 0 0 1 0<br />
Piper guinense Bush pepper Spice 2 3 2 0 1 0 1 1 1<br />
Tetrapleura tetraptera Essisang spice 2 3 1 1 2 1 1 1 1<br />
Fruits/seeds<br />
Dacryodes edulis Plum<br />
Snacks and supplement for meat<br />
(protein)<br />
3 1 1 2 0 0 3 1 1<br />
Cola acuminate 5 piece kola Stimulant and aphrodisiac 3 1 0 1 0 0 1 0 0<br />
Cola nitida 1 piece kola Stimulant and aphrodisiac 0 0 0 0 0 0 2 1 1<br />
Garcinia cola Bitter cola<br />
Stimulant and aphrodisiac, snake<br />
repellent, treatment of coughs and<br />
hepatitis<br />
3 1 0 1 0 0 0 0 0<br />
Aforsthyrix sp. Bush onion Spice 2 1 1 2 0 0 1 0 0<br />
Mushrooms and<br />
vegetables<br />
Agaricus campestries Kokobiako Source of protein and condiment 3 1 1 2 1 0 2 1 0<br />
Gnetum africanum Eru Vegetable 0 0 0 0 0 0 1 2 2<br />
Gnetum buchulzianum Eru Vegetable 0 0 0 0 0 0 1 2 2<br />
Other forest vegetables<br />
(b) – Food of animal<br />
origin<br />
Insects<br />
2 3 2 2 2 1 1 2 1<br />
Apis malifera Honey bee Source of protein 1 3 1 1 3 1 1 2 1<br />
Rhynchophorus<br />
phoenicis<br />
Maggot Source of protein 1 2 1 2 1 2 1 3 1<br />
Bush meat (game)<br />
Cricetomys gambianus Bush rat Source of protein 3 2 1 3 2 2 3 2 1
Table 2. Cond.<br />
Ngane et al. 1791<br />
Thryonomys<br />
swinderianus<br />
Cane rat Source of protein 2 3 2 2 3 2 2 2 1<br />
Cephalopus spp. Duikers Source of protein 3 2 2 2 1 1 2 1 1<br />
Atherurus africanus Porcupine Source of protein 3 2 2 2 3 2 2 3 2<br />
Monkeys Source of protein 3 2 2 3 1 1 3 1 1<br />
Potomochoerus porcus Bush pig Source of protein 2 1 1 2 1 1 1 1 1<br />
Fishes Source of protein 0 0 0 1 1 1 2 1 1<br />
Birds Source of protein 1 1 1 1 1 1 1 1 1<br />
Snakes . Source of protein 3 2 2 3 2 2 2 1 1<br />
Snails<br />
Archachatina. Archatina Snail Source of protein 3 1 1 1 1 1 1 0 0<br />
Archachatina marginata. Snail Source of protein 0 0 0 0 0 0 0 .0 0<br />
0 = none; 1 = very few; 2 = few; 3 = many.<br />
Table 3. Seasonality of major non-food NTFPs in the Kupe area.<br />
NTFP and source Usefulness in the study area<br />
Kupe West<br />
Rainy Dry All year<br />
Kupe North<br />
Rainy Dry All year Rainy<br />
Kupe East<br />
Dry All year<br />
Cordia sp. Construction of houses 3 3 1 3 2 2 2 3 2<br />
Elaeis. Guineensis (palm fronds) Construction of houses 1 3 2 1 3 1 1 3 1<br />
Raphia spp. (palm fronds) Construction of houses 1 2 1 1 2 1 1 2 1<br />
Thaumatococcus spp. (ngongo leaf) Wrapping leaves 3 2 2 2 2 2 3 3 2<br />
Neubouldia laevis Live fencing 3 3 3 2 2 2 1 1 1<br />
Household implements<br />
Bambusa sp. (Indian bamboos) Construction of houses 2 2 2 3 3 3 2 2 2<br />
Cordia sp. Mortar 3 3 1 3 2 2 2 3 2<br />
Elaeis guineensis (palm oil) brooms 1 1 1 1 1 1 2 3 2<br />
Calamus spp. (cane) Can chairs baskets, beds 3 3 3 2 2 2 2 2 2<br />
Acacia spp. (gums) gum 0 0 0 0 0 0 1 1 1<br />
Irvingia spp. carving 1 1 1 1 0 0 0 0 0<br />
Ptericarpus sauyouxii (camwood) Mortar, cosmetics 2 2 2 2 2 2 2 2 2<br />
Firewood fuel 1 3 2 1 2 2 1 2 1<br />
Albizia spp., Irvingia sp. etc.<br />
0 = none; 1 = very few; 2 = few; 3 = many.
1792 Sci. Res. Essays<br />
Table 4. Seasonality of major medicinal NTFPs in the Kupe area.<br />
NTFP and source<br />
Medicine<br />
Croton spp.<br />
Elaeis guineensis<br />
(kernels)<br />
Gnetum africanum<br />
Usefulness in the<br />
study area<br />
Most are marketed:<br />
Chase evil spirits,<br />
stomach ache, well<br />
being of pregnant<br />
women<br />
Babies welfare and<br />
cosmetics<br />
Cure for spleen, bed<br />
wetting, stomach<br />
ache, drunkenness<br />
Rain<br />
y<br />
Kupe West Kupe North Kupe East<br />
Dr<br />
y<br />
All<br />
year<br />
Rainy Dry<br />
All<br />
year<br />
Rain<br />
y<br />
Dry All year<br />
1 1 1 3 3 3 0 0 0<br />
1 1 1 1 1 1 3 3 3<br />
0 0 0 0 0 0 1 2 2<br />
Gnetum buchulzianum 0 0 0 0 0 0 1 2 2<br />
Enanthia chlorantia<br />
(Chloroquine plant)<br />
Azadirachta indica<br />
(Neem)<br />
Garcinia cola (bitter<br />
kola)<br />
Allanblankia floribundia<br />
Prevention and cure<br />
for malaria, added to<br />
local gin<br />
Treatment of malaria,<br />
preservation of<br />
stored crops<br />
Cough, hepatitis,<br />
aphrodisiac<br />
Treatment of<br />
toothaches and<br />
diarrhoea<br />
2 1 1 3 3 3 0 0 0<br />
1 1 1 1 1 1 3 2 2<br />
3 1 0 2 0 0 3 1 1<br />
3 1 1 0 0 0 3 2 1<br />
Albizia sp. (essang) Malaria 2 2 2 2 2 2 1 1 1<br />
Garcinia lucida<br />
(meckak)<br />
Tooth ache 3 3 3 2 2 2 2 2 2<br />
Garcinia mannii (ngou) Tooth bush 3 3 3 2 2 2 2 2 2<br />
Maesopsis eminii<br />
(Rhamnacea)<br />
Afromonum melegueta<br />
(alligator peper)<br />
Cosmetics<br />
Pterocarpus soyauxii<br />
(camwood)<br />
Gonorrhoea and<br />
other bacterial<br />
infections<br />
2 2 2 1 1 1 0 0 0<br />
Charms, aphrodisiac 3 3 3 3 3 2 2 2 2<br />
Rubbed on women<br />
after delivery<br />
3 3 3 2 2 2 2 2 2<br />
Tectona grandis (teak) Bathing sponge 1 1 1 1 1 1 0 0 0<br />
Chewing stick<br />
Garcinia mannii Tooth brush 3 3 3 2 2 2 0 0 0<br />
Massularia acuminata Tooth brush 3 3 3 2 2 2 1 1 1<br />
Animals<br />
Chamelion<br />
(Chamaeleon spp.)<br />
Reduce shock from<br />
fright after accidents<br />
3 3 3 2 2 3 1 1 1
Table 4. Cond.<br />
Pocupine (Atherurus<br />
africanus)<br />
Bush fowl (Francolinus<br />
cameroonensis)<br />
Giant rat (Cricetomys<br />
cambianus)<br />
Grass cutter hair<br />
(Thryonomys<br />
swinderianus)<br />
Easy child birth in<br />
women<br />
0 = none; 1 = very few; 2 = few; 3 = many.<br />
Ngane et al. 1793<br />
2 2 2 2 2 2 1 1 1<br />
Prevent child birth 1 1 1 1 1 1 1 1<br />
Protection against<br />
witchcraft, poisons<br />
and charm to win the<br />
love of women<br />
3 3 3 1 3 2 1 1 1<br />
Healing of eyes 2 2 2 2 2 2 0 0 0<br />
� Table 5. Reasons for seasonality in collection of NTFPs.<br />
�<br />
Reason Kupe East % Kupe North % Kupe west % Total % X<br />
Little farm work 3.3 28.4 38 32.7 45 38.8 11.6 46.4 68.7<br />
Availability 2.0 20.8 34 35.4 4.2 43.7 9.6 38.4 32<br />
Weather 13 34.2 15 39.5 10 26.3 38 15.2 12.7<br />
Total 66 26.4 87 34.8 97 38.8 250 100<br />
Ndume).<br />
Much firewood and yam stakes are collected during the<br />
dry season since land preparation for the next farming<br />
season involves clearing of bushes. Some respondents<br />
revealed that a little more construction material is used<br />
during the dry season. At this time houses and other huts<br />
are constructed using poles, palm frond petioles or<br />
bamboo stems and twines. During the rainy season,<br />
some of these constructions require maintenance<br />
andconstruction materials are again needed to fortify<br />
them.<br />
The study revealed that a particular product could be<br />
obtained from different species in different zones. For<br />
instance, palm wine is obtained from both Elaeis<br />
guineensis and Raphia hookery in all the clans, but the<br />
latter is more commonly found in fresh water swamps in<br />
Kupe West and more available in the rainy season as it is<br />
easier to tap (less readily diluted by rain water during<br />
tapping).<br />
Though collection of NTFPs provides employment to<br />
the inhabitants of the area, agriculture is the mainstay of<br />
the people of Kupe. The people are more involved in<br />
farming activities during the dry season than during the<br />
rainy season. Little time is spent on gathering of NTFPs<br />
during the period of crop cultivation. However, during<br />
agricultural slack periods the collection, processing and<br />
marketing of NTFPs becomes the major occupation of<br />
some rural dwellers. More fulltime employment was<br />
observed in Kupe West and Kupe North than in Kupe<br />
East (Figure 3). This is probably because of the<br />
abundance of the products in Kupe West and Kupe North<br />
than in Kupe East (Tables 2, 3, and 4).<br />
CONCLUSION<br />
This work has thrown some light on the availability and<br />
collection of non-timber forest products in Kupe. These<br />
products show seasonality in availability and collection<br />
with most of them being collected during the rainy season<br />
when there is little work in the farms and when food crops<br />
are yet to mature. The duration of the rainy season<br />
(7months) in the study area is more than that of the dry<br />
season (5 months). This is another reason why more<br />
NTFP collection takes place during the raining season.<br />
Collection of NTFPs is an income diversification activity in<br />
The area and is also a risk-reducing strategy against<br />
household income fluctuations.<br />
It was observed that during particular periods of the<br />
year a particular NTFP will exist in abundance and<br />
sometimes in excess while there is acute scarcity during<br />
another period of the year. This seasonal nature of<br />
NTFPs has been an enormous setback in the utilisation<br />
of these important resources. To enable the local people<br />
and other actors benefit from the exploitation of NTFPs,<br />
research on production of off-season varieties should be
1794 Sci. Res. Essays<br />
Atop<br />
Ngusi<br />
Kupe<br />
West<br />
… Mount Kupe farm/forest boundary<br />
… Clan boundary<br />
� Study villages and forest<br />
Mpako<br />
Tape-<br />
Etube<br />
Bendume<br />
Mount<br />
Kupe<br />
Figure 1. Location map of the study site (Mt. Kupe area).<br />
Figure 1. Location map of the study site (Mt. Kupe area).<br />
encouraged. Selection for all-year-round production for<br />
important NTFPs such as D. edulis and Irvingia spp. can<br />
certainly have considerable impact on the availability of<br />
Nyasoso<br />
Nlohe<br />
Kupe<br />
East<br />
Kola<br />
Lala<br />
products as there is typically some within-species<br />
variation in phenology with some trees flowering and<br />
fruiting outside the main season. Artificial and deliberate
Percentage of respondents<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
0<br />
Seasonality of NTFP Collection in the Kupe Area<br />
Kupe East Kupe North Kupe West<br />
Zone<br />
�<br />
� Figure 2. Seasonal variation of NTFP collection in the study area.<br />
Percentage of respondents<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
0<br />
Employment pattern of NTFPs<br />
Part-time Full-time Part-time Full-time<br />
Rainy season Dry season<br />
Pattern and seasons<br />
Rainy<br />
Dry<br />
All year<br />
K West<br />
K North<br />
K East<br />
Figure 3. Effect of geographical zone and different seasons of the year on employment patterns in<br />
the survey zone.<br />
actions such as improved preservation techniques,<br />
irrigation and incorporation of NTFPs in agroforestry<br />
systems and organic gardens can be important in<br />
Ngane et al. 1795<br />
providing off-season products. This will be very useful<br />
today in the mitigation of the effect of climate change on<br />
availability, collection and management of NTFPs.
1796 Sci. Res. Essays<br />
A B C<br />
D E<br />
G H<br />
J K L<br />
Figure 4. Some common NTFPs in the Mount Kupe area found in the Tombel Town Market place.<br />
REFERENCES<br />
Alika JE (1997). Statistics and Research Methods. 1 st Edition, Ambik<br />
Press Publishers, Benin City, Nigeria.<br />
Biliso A, Lejoly J (2006). A Study of the Exploitation and Markets of<br />
Non-Timber Forest Products in Kinshasa. Tropicult, 24(3): 183-188.<br />
Buyungu PM (2000). The Contribution of Plums (Dacryodes edulis) in<br />
the diversification of Agriculture in West Africa. In kengue J, Capseu<br />
C, Kayem GJ (eds.), Third International Seminar on the Valorisation<br />
of Plums and other Non-Conventional Oil Crops. Yaounde, 3-5<br />
Octobre 2000. DFID, CTA IFS CIRAD, Pub. African University Press<br />
Yaounde, 638 p. Programme (E.D.A.D.P.). Ministry of Agriculture,<br />
Benin City. Edo State.<br />
Dijks Van (1999). An Assessment of Non-wood Forest Product<br />
Resources for Development of Sustainable Commercial Extraction,<br />
In: Sunderland TCH, Clark LE, Vantomme P (Eds.) Non-wood forest<br />
F<br />
I<br />
products of Central Africa: current research <strong>issue</strong>s and prospects for<br />
conservation and development. Based on the outcome of the<br />
International Expert Meeting on Non-wood Forest Products in Central<br />
Africa, held at the Limbe Botanic Garden, Cameroon, 10-15 May<br />
1998. FAO, URL: htt://www.fao.org/do crep/x216 le 32. Htm .<br />
Ejedepang-Koge SN (1986). The Tradition of a People Bakossi: A<br />
Historico-Socio Anthropological Study of one of Cameroon’s Bantu<br />
Peoples. Virginia, USA: ARC Publications, 362: 42-45.<br />
Falconer J (1992). Non-Timber Forest Products in Southern Ghana.<br />
Wilson et al., (eds). Publ. National Resources Institute, Chatham M.<br />
K. ME4 4TB, UK., pp. 6-7.<br />
Food and Agricultural Organisation of the United Nations (FAO) (1991).<br />
Marketing Non Wood Forest Products in Developing Countries.<br />
UNASYL. J., 46(1): 37-41.<br />
Hannagan TJ (1989). Mastering Statistics. 2 nd edition, MacMillan<br />
Education. Ltd, 272: 43-57.
Hart AD, Azubuike CU, Barimala IS, Achinewhu S C (2005). Vegetable<br />
consumption pattern of households in selected areas of the Old<br />
Rivers State in Nigeria. Afri. J. Food, Agric. Nutri. Dev., 5(1): 18-25.<br />
Kio PRO, Abu JE (1993). Environmental accounting mechanism for<br />
reconciling land use pressure on forests; In: PRO Kio. (Ed),<br />
UNEP/CIFOR Project for West/Central Africa, University of Ibadan,<br />
Ibadan, Commer. For. Rev., 72(4): 272-278.<br />
Leakey RR, Greenwell BP, Hall MN (2000). Domestication of<br />
indigenous fruit trees in West and Central Africa: In: kengue J,<br />
Kapseu C, Kayem GJ (eds.), Third International Seminar on the<br />
Valorisation of Plums and other Non-Conventional Oil Crops.<br />
Yaounde, 3-5 Octobre 2000. DFID, CTA IFS CIRAD, Pub. African<br />
University Press Yaounde, 638: 73-92.<br />
Ngane et al. 1797<br />
Ndoye O, Tieguhong JC, (2004). Forest Resources and Rural<br />
Livelihoods: The Conflict between Timber and Non-timber Forest<br />
Products in the Congo Basin. Scan. J. Forest. Res., 19: 36-44.<br />
Ngwasiri C, Djeukam N, Vabi M (2005). Lagislative and Institutional<br />
Instruments for the Sustainable Management of Non-timber Forest<br />
Products (NTFPs) in Cameroon. Past, Present and Unresolved<br />
Issues. Community Forestry Development Project, Yaoundé,<br />
Cameroon.<br />
Okafor JC (1989). Agroforestry Aspects. Appendix 2 of the Cross River<br />
National Park Oban Division Plan for Developing the Park and Its<br />
Support Zone. Godalming. W.W.F, U.K, pp. 7-90.
Scientific Research and Essays Vol. 7(18), pp. 1798-1808, 16 May, 2012<br />
Available online at http://www.academicjournals.org/SRE<br />
DOI: 10.5897/SRE12.223<br />
ISSN 1992-2248 ©2012 <strong>Academic</strong> <strong>Journals</strong><br />
Full Length Research Paper<br />
Adaptive message size routing strategy for delay<br />
tolerant network<br />
Qaisar Ayub, M. Soperi Mohd Zahid*, Sulma Rashid and Abdul Hanan Abdullah<br />
Department of Computer System and Communication, Faculty of Computer Science and Information System,<br />
Universiti Teknologi Malaysia (UTM) Skudai - Johor, 81310, Malaysia.<br />
Accepted 3 May, 2012<br />
Delay tolerant network (DTN) is a kind of computer network that suffer from the frequent<br />
disconnections, network partitioned and unstable network connectivity, therefore maintaining an<br />
uninterrupted route from source to destination is not possible. Therefore, the transmission of message<br />
is achieved via intermediate nodes by adopting a novel transmission mechanism called store-carry and<br />
forward where node stores the incoming message in its buffer, carries it while moving and forward it<br />
when it comes in the transmission range of other nodes. DTN routing protocols can be either single<br />
copy or multi copy. In single copy protocols, the node forwards the unique copy of message along a<br />
single path. These protocols suffer the long delivery delay. In multi copy protocols, the node diffuses<br />
multiple copies of same message along dissimilar paths. Thus, message can reach destination via more<br />
than one path. However, the replication process consumes high volume of network resources such as<br />
buffer space, bandwidth and node energy. The probabilistic routing strategies for instance PRoPHET<br />
Protocol minimizes the consumption of resources and forwards a message to a custodian by using a<br />
metric of delivery probability. The probability describes the suitability of a node to meet the destination<br />
of message. However, when node mobility pattern is not symmetric the probabilistic computations<br />
cannot predict the accurate forwarding decision. In this paper, we have proposed a novel message<br />
forwarding strategy called Adaptive Message-Size Routing strategy (AMRS) by which a node handovers<br />
the copy of message to its neighboring nodes by using a metric named as mean threshold (MTH). We<br />
have compared the performance of AMRS with Epidemic and PRoPHET routing protocols. The<br />
proposed routing strategy has performed better in terms of maximizing delivery probability while<br />
minimizes message drops and number of transmissions.<br />
Key words: Store-carry-forward, routing protocols, delay tolerant network, algorithm.<br />
INTRODUCTION<br />
Ad hoc routing protocols for instance table driven and ondemand<br />
(Johnson et al., 2001; Ahmedy et al., 2011)<br />
establish an end-to-end path prior to the transmission of<br />
data. However, these solutions are not applicable in<br />
applications like wild life monitoring, deep space<br />
*Corresponding author. E-mail: soperi@utm.my.<br />
Abbreviations: MTH, Mean threshold at active node; BU,<br />
buffer used by the active node; Mn, Number of messages at<br />
active node; MDest, message destination; Msize, message size;<br />
APN, Appetizer Node; ACN, active node.<br />
communication, military and sensor network where the<br />
node mobility causes dynamic topology changes and<br />
frequent disconnections. As a result, source and<br />
destination cannot sustain the uninterrupted path.<br />
The delay tolerant network (Fall and Farrell, 2008)<br />
provides the transmission of data for such scenarios by<br />
applying the novel message diffusion strategy referred to<br />
as store, carry and forward. According to this approach,<br />
the node stores the incoming message in its buffer,<br />
carries it while moving and forwards when comes in the<br />
transmission range of other nodes. These networks are<br />
also known as opportunistic because the node always<br />
searches for a connection opportunity to forward its<br />
buffered messages. The connection happens when
node(s) move in the communication range of each other.<br />
These transmissions continue and a message finds its<br />
destination via multiple hop(s). Based on this paradigm,<br />
various delay tolerant network (DTN) routing protocols<br />
have been proposed to cope with the heterogeneous<br />
network scenarios. The available routing strategies can<br />
be categorized as opportunistic, predicted and<br />
scheduled. The opportunistic protocol requires least<br />
computational resources and node simply forwards the<br />
message copy to the encounter nodes for example,<br />
Epidemic (Vahdat and Becker, 2000), Spray and wait<br />
(Thrasyvoulos et al., 2005). The predicted protocol<br />
forwards the message by utilizing the contact history of<br />
current node such as PRoPHET (Lindgren et al., 2003).<br />
While the scheduled based forwarding transmits the<br />
message by employing the human mobility movement<br />
pattern of contacts and positional coordinates, in DTN<br />
each node is hybrid, which means that it can generate<br />
and receive the random size messages. Hence, in the<br />
presence of limited bandwidth the small size messages<br />
swiftly pass through the network. This increases the<br />
message replication that results in rapid clogging on the<br />
buffer space of network nodes. Additionally, when a node<br />
by carrying small size messages receives the large size<br />
message it will drop the cluster of messages to<br />
accommodate new arrival. As, there exists multiple<br />
copies of each message, therefore same node may<br />
receive the dropped messages from other part of<br />
network. These rampant drop and re-transmissions<br />
wastes the bandwidth, energy and formulates high<br />
burden on the network resources. Hereby, a mechanism<br />
is required which control the generation of message<br />
copies and reduces the number of message drops.<br />
In this paper, we have proposed a new message<br />
routing strategy called Adaptive Message-Size routing<br />
strategy (AMRS),where a node handover the copy of<br />
message to its neighbors by employing a metric Mean<br />
Threshold (MTH). According to AMRS, before trans-<br />
mitting the message ‘M’ to ‘C’, the node ‘A’ computes the<br />
mean of buffered messages at ‘C’. If the size of message<br />
‘M’ is less than MTH of ‘C’ and ‘M’ is not the destination<br />
for ‘C’ then the ‘A’ will keep such messages in its buffer.<br />
However, ‘A’ will forward ‘M’ to ‘C’ only if message is<br />
destine for ‘C’ or the size of message is greater then<br />
MTH of ‘C’.<br />
We have compared the performance of AMRS with<br />
Epidemic and PRoPHET Protocols. Under all simulation<br />
configurations the AMRS always reduce the transmission<br />
overhead, message drop and raises delivery probability.<br />
RELATED WORKS<br />
DTN routing protocols attempts to transfer the data over<br />
the unstable communication. Thus, transmission of<br />
multiple message copies along different paths makes<br />
sure that at least one copy will reach the destination. An<br />
extreme example of such policy is the Epidemic (Vahdat<br />
Ayub et al. 1799<br />
and Becker, 2000) routing. In this protocol, when two<br />
nodes meet, they exchange all messages which they do<br />
not have in common. In this way, message(s) rapidly<br />
diffuses in the network and can reach the destination via<br />
multiple paths. This increases the robustness, minimize<br />
the delivery delay but consumes high volume of network<br />
resources such as bandwidth, buffer space and node<br />
energy. Some recent work (Qaisar et al., 2011) has<br />
optimized the performance of Epidemic routing by<br />
changing the default transmission order FIFO. The spray<br />
and wait routing (Thrasyvoulos et al., 2005) reduces the<br />
consumption of network resources by minimizing the<br />
transmission of message copies. In this scheme, the<br />
source node forward the ‘N’ number of its message<br />
copies to the encounter nodes called relays. This is spray<br />
phase. If the destination is not found in spray phase, then<br />
each node wait and forwards the message directly to<br />
destination. This protocol delivers the message like<br />
Epidemic fashion while it minimizes the consumption of<br />
network resource like direct delivery. Spray and focus<br />
(Thrasyvoulos et al., 2007, 2008) is another deviation of<br />
spray and wait where the node starts by spraying the N<br />
number of message copies. If the message does not find<br />
the destination then the node hands over the message to<br />
a custodian with higher utility. The utility value determines<br />
capability of node to meet the message destination.<br />
The contribution in routing protocol continues and the<br />
researcher begins to route the message by exploiting the<br />
additional network parameters such as the mobility<br />
pattern, encounter history activity of node. For example,<br />
the PRoPHET routing protocol (Lindgren et al., 2003)<br />
utilizes the encounter history and age of last encounter.<br />
Moreover, it route a message by using the metric of<br />
delivery predictability. NECTAR (Etienne and de Oliveira,<br />
2009) protocols controls the transmission of messages by<br />
computing the neighborhood index, which depends on<br />
the encounter history. Srinivasa and Krishnamurthy<br />
(2009) proposed a distributed proximity-based<br />
communication protocol that forwards the message using<br />
‘conditional residual’ time metric. This metric uses the<br />
local knowledge of past contact to estimate remaining<br />
meeting time for pair of nodes. CREST protocol<br />
minimizes the end-to-end delay and increase the delivery<br />
probability as compared to other message forwarding<br />
strategies MEED and MED.<br />
Conditional Shortest Path protocol (CSPR) (Eyuphan et<br />
al., 2010) effectively employs the human mobility pattern<br />
and computes the conditional intermitting time, which is<br />
the average meeting time of two nodes in relation to<br />
meeting time with third node. The average intermitting<br />
time represents the link cost. The node forwards the<br />
message by employing conditional shortest path<br />
algorithm that takes conditional average intermitting time<br />
as parameter. CSPR minimize the end-to-end delay and<br />
increase the delivery probability as compared to shortest<br />
path based routing protocols. Guizhu et al. (2010)<br />
improves the performance of Binary spray and wait<br />
algorithm by introducing QoN. The QoN is measured in
1800 Sci. Res. Essays<br />
Figure 1. Appetizer node.<br />
terms of mobility of node and is represented by an integer<br />
number which indicates how frequent one node<br />
encounter with other node in a given time interval. Higher<br />
value of encounters indicates the high QoN. Messages<br />
are forwarded to a node that has higher value of QoN.<br />
When a node leaves one copy, it then switches to direct<br />
transmission. QoNsW increase the delivery probability,<br />
minimize the delivery delay, overhead as compared to<br />
Epidemic and spray and wait.<br />
PROCS (Jathar and Gupta, 2010) minimize the<br />
replication of message by adopting a forwarding strategy,<br />
which study the movement pattern of contacts and their<br />
time sequence. In PROCS, each node maintains a<br />
probabilistic contact graph that represents the contact<br />
probability among nodes at various time intervals. The<br />
node route the message by selecting a contact with<br />
greedy path calculation. PROCS protocol achieves high<br />
message delivery ratio with no message replication<br />
compared to Epidemic and PRoPHET. Daowen et al.<br />
(2009) manages the routing computation locally at each<br />
and transmits the message by using hierarchical<br />
forwarding with cluster control mechanism. This protocol<br />
split the network into different clusters. A node called<br />
cluster head controls each cluster. The node first sends<br />
the request to forward the message towards cluster head.<br />
The cluster head analyze the message header and<br />
perform Inter-cluster or Intra cluster forwarding. CHRC<br />
reduce the delivery delay and improve the delivery<br />
probability as compare to Epidemic and SMART<br />
protocols. The only drawback of CRHC is that it needs to<br />
maintain the information regarding partial nodes of the<br />
network.<br />
The proposed Adaptive Message-Size forwarding<br />
strategy forwards a message without exploiting and<br />
relegating the complex computations. The idea is to<br />
deliver or relay the message, the mobility of node and<br />
observing the buffer of encountered terminal. For this, a<br />
metric Mean Threshold (MTH) have been defined which<br />
controls the creation of multiple copies of each message.<br />
AMRS routing Algorithm (adaptive message size<br />
forwarding strategy)<br />
Terminology<br />
In proposed AMRS, each node maintains the index of its<br />
buffered messages. The node with high volume index<br />
controls the communication and is known as appetizer<br />
node. The volume index is determined in terms of<br />
number of message carried by a node. When two nodes<br />
have the same volume index then the selection of<br />
appetizer is random. The nodes ready to communicate<br />
with appetizer are known as active nodes. A node is<br />
active if it is not busy in communicating with its<br />
neighbors.<br />
The appetizer node controls the flow of messages<br />
around the active connections. For this, appetizer fetches<br />
its buffered message one at a time and transmits it by<br />
verifying the Rule 1, 2. After the ‘appetizer’ <strong>complete</strong>s its<br />
transmission, it changes its status to active node while<br />
the next high volume index node becomes the ‘appetizer’.<br />
Figure 1 represents ‘A’ as appetizer while ‘B’, ‘D’ and ‘E’<br />
are the active nodes.<br />
METHODOLOGY<br />
Rule 1: “The appetizer ‘A’ will forward message ‘M’ to ‘B’ only<br />
if ‘B’ is destination of ‘M’ and ‘B’ has not previously received<br />
the same message.”<br />
The existing Epidemic protocol (Vahdat and Becker, 2000) is a<br />
multi copy routing strategy where on each encounter, the node<br />
perform the pairwise exchanges of messages. Theses nodes<br />
become the carrier and continue to infect the further encountering<br />
terminals. In this way, the message rapidly moves across the<br />
disconnected regions.<br />
Despite delivering the messages, this protocol leaves high<br />
overhead on the network resources for instance buffer space, node<br />
energy and bandwidth. The previous work (Vahdat and Becker,<br />
2000) has taken the buffer space, bandwidth as infinite resources<br />
that is not practical for the real time DTN applications. For example,<br />
when the bandwidth is limited, then the node may not fully<br />
exchange their buffered messages and may miss the transmission<br />
of those messages for which the current connected node is the final<br />
destination.<br />
The message delivery to the destination is very important as<br />
when a message reach the recipients as a final custodian then that<br />
node generates an acknowledgement message. This message<br />
informs the other nodes about the status of delivered message and<br />
operates as a sweeper to remove the message copies and stop the<br />
further replication. However, when messages doest not reach the<br />
destination, their replication continues in the network.<br />
Rule 1 of the algorithm ensures the delivery of more messages to<br />
their destinations. For this, the appetizer node scans its buffer one<br />
message at a time and validates rule 1. Hence, the transmission<br />
occurs only if ‘M’ is destined for current connected node. However,<br />
when the message is not destined for ‘B’ then Rule II invokes.<br />
Rule 2: “The appetizer will not forward ‘M’ to ‘B’ only when the<br />
size of message ‘M’ is less than the mean of buffered<br />
messages at MTH (B).”<br />
In DTN, each node is capable of generating or receiving random
for each ACN in communication range of APN<br />
{ MTH (ACN) = BU/Mn;<br />
for each message ‘M’ in APN<br />
{<br />
if (Mdest==ACN)<br />
forward_message (M,ACN);<br />
else if (Msize < MTH (ACN))<br />
Continue;<br />
else<br />
Forward_message (M,ACN);<br />
}<br />
}<br />
Figure 2. Algorithm (AMRS).<br />
size messages. Further, these messages reach their destination via<br />
multiple intermediate hops called relays. Since, message relay is<br />
directly proportional to the congestion. The high number of<br />
message relays increases the network traffic. While under the<br />
limited buffer space, the nodes could not accommodate the allincoming<br />
transmissions. This congestion results in dropping the<br />
previously stored messages. As, there exists multiple copies of<br />
each message therefore same node may receive the dropped<br />
messages from other part of the network. These rampant drops and<br />
re-transmissions waste the bandwidth, energy and formulates high<br />
burden on the network resources.<br />
One factor that can be used to control the congestion is to<br />
minimize the number of message relays. Rule 2 deals with<br />
controlling the unnecessary relay of messages. For this, before<br />
forwarding the message ‘M’, appetizer ‘A’ computes the mean of<br />
buffered messages at ‘B’ which is called Mean Threshold (MTH).<br />
The mean value indicates the average size of message carried by<br />
the active node. If the size of ‘M’ at ‘A’ is less than the MTH (B),<br />
then it shows that as compared to ‘B’ the current message ‘M’ at ‘A’<br />
is small. In addition, ‘B’ is not the destination of ‘M’ therefore<br />
relaying the message to ‘B’ does not guarantee its eventual<br />
delivery. At this stage, both nodes are equally capable to meet with<br />
the destination. Therefore relaying the message is waste of node<br />
energy, bandwidth and buffer space. Thus, ’A’ switch to the greedy<br />
mode and carry such small size messages in its buffer.<br />
As, all network nodes has random size messages. Therefore,<br />
MTH value varies for each node. Hence, appetizer adaptively will<br />
not relay small size messages and the transmission occurs only<br />
when size of message ‘M’ is greater than the MTH of active node<br />
(Figure 2).<br />
Message exchange<br />
Case 1: Exchange of Messages Epidemic Protocol<br />
Consider a sample scenario in Figure 3 where nodes ‘A’ and ‘B’ has<br />
established the connectivity while SVA, SVB represent the buffered<br />
messages at A and B. Hence, SVA = { (m1, 39 s, X, 200 KB), (m2,<br />
53 s T, 150 KB) ,(m3, 46 s, U, 180 KB), (m4, 22 s, B, 230 KB), (m5,<br />
10 s, B, 153 KB) | (message, arrival time, destination, size)}, SVB={<br />
(m50, 41 s, X, 150 KB), (m51, 32 s, Z, 410 KB), (m52, 78 s, M, 200<br />
KB), (m53, 15 s, Q, 240 KB), | (message, arrival time, destination,<br />
size)}.<br />
According to Epidemic routing protocol, ‘A’ forwards its summery<br />
Ayub et al. 1801<br />
Figure 3. Exchange of messages using Epidemic routing protocol.<br />
vector to ‘B’. Node ‘B’ then computes the summery vector request<br />
SVR by subtracting the SVA from its own SVB to get the messages<br />
not buffered at ‘B’.<br />
The node ‘A’ then forwards the required messages to ‘B’ by<br />
arranging them according to the arrival time that is, the message<br />
with high arrival time should be placed on the top of the queue such<br />
as SV(A-B) = {m2, m3, m1, m4, m5}.<br />
Since bandwidth is the limited resource therefore it is possible<br />
that ‘A’ may not be able to forward m4, m5 that are placed at the<br />
end of the forwarding queue. Both m4 and m5 were destined for the<br />
current connection ‘B’. Therefore, instead of delivering the<br />
messages, ‘A’ will continue the replication m4, m5 on the other part<br />
of the network.<br />
Case 2: Exchange of messages via Adaptive Message Size<br />
forwarding strategy<br />
Now we will consider the transmission of messages via adaptive<br />
message size forwarding strategy. Consider the snapshot<br />
mentioned in Figure 4; here, ‘A’ act as appetizer because it has<br />
high volume index compared to ‘B’. Recall that appetizer node fetch<br />
its buffered messages one by one and forward it to the current
1802 Sci. Res. Essays<br />
Figure 4. Exchange of messages using AMRS.<br />
active node by Rule 1 and Rule 2.<br />
In present scenario, appetizer ‘A’ will fetch ‘m1’ from its buffer<br />
and validate it by Rule 1. As m1 is not destined for ‘B’, thus by Rule<br />
2 ‘A’ will obtain the mean of the buffered messages MTH at ‘B’ that<br />
is MTH (B) = 250 KB. Since, the size of message m1 is less than<br />
the MTH (B) thus ‘A’ will keep ‘m1’ in its buffer and move to the next<br />
message ‘m2’. Similarly, ‘m2’ is not destined for ‘B’. In addition the<br />
size of ‘m2’ is also less than mean threshold MTH(B) thus ‘A’ keep<br />
this message in the buffer and moves to the next message ‘m3’ that<br />
will not be transmitted because it does not satisfy the algorithm<br />
rules. Thus, appetizer ‘A’ iterate the next message ‘m4’ which is<br />
destined for current connected node ‘B’; therefore ‘A’ will forward<br />
this message to ‘B’ and iterate the next message ‘m5’ that is<br />
destined for ‘B’; and ‘A’ will transmit ‘m5’ to ‘B’.<br />
SIMULATION AND RESULTS<br />
Here we will compare the performance of proposed<br />
Adaptive message size forwarding strategy (AMRS) with<br />
Epidemic and PRoPHET routing protocols by varying the<br />
various simulation parameters. The evaluation of<br />
schemes have been analyzed under ONE [22] simulator.<br />
ONE is a discrete event simulator written in JAVA and<br />
have been massively used by various researchers to<br />
measure the statistics of disrupted store-carry-forward<br />
applications.<br />
Network settings<br />
The simulations were performed by considering a citybased<br />
environment consisting two groups of pedestrians<br />
(80 nodes), one group of car (40 nodes) and six trains.<br />
The pedestrians and cars are moving according to<br />
Shortest Path Map Based Movement mobility model<br />
while trains are moving with Map Route Movement<br />
mobility model. The pedestrians were configured at the<br />
speed of 0.5 to 1.5 km/h for pedestrians, 10 to 50 km/h<br />
for cars and 10 to 50 km/h for trains with the transmission<br />
range of 10 m. We have configured each network peer<br />
with a finite buffer space and limited bandwidth 2 MBPS.<br />
In addition, the connection establishment is opportunistic<br />
and nodes do not have the knowledge about the network<br />
topology.<br />
Performance metrics<br />
Message relays<br />
DTN is an opportunistic network where messages reach<br />
its destination via multiple intermediate hop(s) referred to<br />
as relays. When a message is relayed, it consumes the<br />
energy, buffer space and bandwidth. The redundant<br />
diffusion of messages puts high load on them. Therefore,<br />
it is important to minimize the number of transmissions or<br />
relays.<br />
Message dropped<br />
Since, the transmission of multiple message copies<br />
produces high congestion on the buffer of intermediate<br />
node(s), in result, the node overcomes it by dropping<br />
previously stored messages. It is not possible to remove<br />
the drop event; however reducing its magnitude can<br />
improve the network throughput.<br />
Delivery probability<br />
The delivery probability measures the successful<br />
transmission of messages to their destinations. This<br />
metric measures the overall network throughput, as more<br />
messages delivery at destination shows the optimal use<br />
of network resources.<br />
Performance evaluation<br />
Scenario-01: Impact of varying number of nodes on<br />
routing protocols<br />
Figure 5 represents the results of PRoPHET, Epidemic
Figure 5. Message relays by varying number of nodes.<br />
Figure 6. Message drop by varying number of nodes.<br />
and AMRS in term of message relay. We can see that<br />
Epidemic protocol has the highest number of relays. The<br />
reason is that the message exchange rate of Epidemic<br />
protocol depends on the number of encounter. Thus, at<br />
higher nodes density such as 246, 216 and 186 the<br />
encounter rate among nodes also increases which in turn<br />
increase the maximum transmissions.<br />
The PRoPHET protocol has relayed fewer messages<br />
than Epidemic protocol. The reason is that PRoPHET<br />
Protocol does not relay the message if the encountered<br />
node is less probable to meet the destination. However,<br />
when we increase the node density for example [186],<br />
[216], [246] the PRoPHET protocol begins to relay more<br />
Ayub et al. 1803<br />
messages. The reason is that the PRoPHET protocol<br />
uses the history of encounters to control the replication of<br />
messages, in which high density of nodes elevates the<br />
encounters rate. In response more nodes has maintained<br />
the predictability measure for each other.<br />
The proposed AMRS shows the consistent message<br />
transmissions. The reason is that, AMPR generates the<br />
message copy only if the size of current message is<br />
greater than the MTH of connected node. In this way all<br />
nodes carry the small size messages and tries to deliver<br />
it directly to destination.<br />
Figure 6 shows the results of message drop in term of<br />
number of nodes. We can observe that the Epidemic
1804 Sci. Res. Essays<br />
Figure 7. Delivery probability by varying number of nodes.<br />
protocol has dropped the high number of massages. The<br />
reason is that the Epidemic protocol blindly floods the<br />
message copies. While, under limited buffer space the<br />
encountering terminals could not accommodate all<br />
incoming traffic.<br />
The PRoPHET Protocol controls the replication of<br />
messages; therefore has dropped less messages than<br />
Epidemic. The AMRS reduces the message drop as<br />
compared to Epidemic and PRoPHET routing protocol.<br />
The reason for this is that, the Mean Threshold (MTH)<br />
moves the traffic to the less congested part of the<br />
network.<br />
Figure 7 depicts the result of delivery ratio for AMRS,<br />
PRoPHET and Epidemic routing protocols with respect to<br />
the number of nodes. We can see that the Epidemic<br />
protocol delivered the least number of messages. The<br />
reason is that, Epidemic protocol does not control the<br />
replication of messages as a result produce congestion<br />
on the network. Added together, protocol also overcomes<br />
this problem by dropping its carried messages.<br />
Therefore, high number of messages dropped before<br />
finding the destination. The PRoPHET protocol delivers<br />
more messages than Epidemic protocol. The reason is<br />
that PRoPHET protocol controls the replication of<br />
messages as well as forwards the message by<br />
computing the delivery probability.<br />
The proposed AMPR routing protocol double the<br />
delivery of messages for all simulation instances.<br />
However, at 246, the protocol has delivered maximum<br />
number of messages. The reason is that AMRS does not<br />
generate and forwards the copy of a small size message.<br />
This in turn, minimizes congestion, less number of drops<br />
and increase the message stay time in the buffer, which<br />
results in high delivery probability.<br />
Scenario 2: Impact of varying buffer size on routing<br />
protocols<br />
Figure 8 maps the results of AMRS, PRoPHET and<br />
Epidemic routing protocols in terms of message relays.<br />
We can observe that, as we increase the storage<br />
capacity such as 4 M, 5 M, 6 M, existing Epidemic and<br />
PRoPHET Protocol has increased the message relays.<br />
This is because message relay depends on the storage<br />
capacity of encounter node.<br />
The proposed (AMRS) routing strategy has maintained<br />
a consistent number of relays. The reason is that AMRS<br />
relays the message copy by computing the MTH and thus<br />
does not depend on the buffer size.<br />
Figure 9 shows the result of message drop for existing<br />
PRoPHET, Epidemic and proposed AMRS. We can see<br />
that for all simulation instances Epidemic protocol has<br />
high drop ratio. Moreover, as the storage capacity rises<br />
such as 4 M, 5 M, 6 M the drop also increases. The<br />
reason is that, Epidemic protocol relays high volume of<br />
messages that results in congestion. Thus, each node<br />
iteratively drops the previously stored messages to<br />
overcome the congestion and continue the flow of<br />
network traffic. The PRoPHET routing protocol controls<br />
the replication of message thus drop less messages than<br />
Epidemic. However, the proposed AMRS has dropped<br />
least amount of messages. The reason is that, the<br />
proposed strategy (AMRS) has minimum number of<br />
message relays.<br />
Figure 10 depicts the results of existing PRoPHET,<br />
Epidemic and proposed AMRS routing protocols in terms<br />
of delivery probability. We can see that with increase in<br />
the buffer size for example 3 M, 4 M, 5 M, and 6 M all<br />
routers has performed better. Moreover, AMRS routing
Figure 8. Message relay by varying buffer size.<br />
Figure 9. Message drop by varying buffer size.<br />
protocol has reciprocated well for all simulation instances<br />
and delivers messages two times higher than the existing<br />
strategies. The reason is that the delivery of message<br />
depends on the buffer time of the message; a message<br />
with high buffer stay time is more likely to be delivered to<br />
its destination.<br />
Scenario 3: Impact of varying message size on<br />
routing protocols<br />
Figure 11 represents the results of number of message<br />
relays for existing PRoPHET, Epidemic and proposed<br />
AMRS routing protocols by varying the sizes. We can<br />
Ayub et al. 1805<br />
evict that at the range of small size messages for<br />
instance [100 K – 600 K], [200 K – 700 K] both Epidemic<br />
and PRoPHET has the maximum number of<br />
transmissions. The reason is that the small size<br />
messages can float around the network more quickly. In<br />
addition, we can see that at large size messages for<br />
example [300 K -800 K], [400 K – 900 K] and [500 K – 1<br />
MB] all routing protocols has reduced the message<br />
relays.<br />
The proposed AMRS does not depend on the range of<br />
message size. We can observe that for all simulation<br />
instances the proposed protocol has shown a constant<br />
number of message relays.<br />
Figure 12 maps the finding of message drop by varying
1806 Sci. Res. Essays<br />
Figure 10. Delivery probability by varying buffer size.<br />
Figure 11. Message relay by message size.<br />
the size of messages. We can observe that at small<br />
range of message sizes such as [100 K – 400 K], [200 K<br />
– 700 K], [300 K – 800 K] both Epidemic and PRoPHET<br />
has higher number of message drops. The reason is that<br />
small size messages tends to infect the network more<br />
rapidly and produces the congestion.<br />
Thus, nodes begin to drop the previously stored<br />
messages. In contrast, at large range for example [400 K<br />
– 900 K], [500 K – 1 MB] fewer messages were dropped.<br />
However, in case of AMRS, each node tends to carry the<br />
small size messages.<br />
Figure 13 conclude the ratio of successful transmission<br />
of messages to their destinations under the variable size<br />
of messages. In case of existing Epidemic and PRoPHET<br />
routing protocols, when the message range is small in<br />
size for example [100 K - 600K], [200 K – 700 K] more<br />
messages reaches the destination. The reason is that the<br />
small size messages are likely to find destination more<br />
quickly. However, the proposed routing protocol AMRS<br />
has delivered high number of messages to their<br />
destinations. The protocol has shown better results when<br />
there is high congestion in the network and message size<br />
is very small.<br />
Conclusion<br />
In this paper, we propose a new routing strategy known<br />
as adaptive message-size forwarding strategy (AMRS).<br />
According to the proposed strategy, a node generates
Figure 12. Message drop by random message size.<br />
Figure 13. Delivery probability by random message size.<br />
and handover the copy of message to neighboring nodes<br />
by using a metric named as Mean Threshold (MTH). We<br />
compared the performance of AMRS with Epidemic and<br />
PRoPHET Protocols. The proposed strategy has<br />
maximized the delivery probability while it controls the<br />
message drops and number of transmissions.<br />
REFERENCES<br />
Ahmedy I, Asri NMD, Syaril NOM, Junaid C (2011). A review on<br />
wireless sensor networks routing protocol: Challenge in energy<br />
perspective. Sci. Res. Essays, 6(26): 5628-5649.<br />
Daowen Hua, Xuehui Du, Yanbin Qian, Shaoge Yan (2009). Routing<br />
Protocol Based on Hierarchy Forwarding and Cluster Control. CIS’,<br />
pp. 397-401.<br />
Ayub et al. 1807<br />
Etienne C, de Oliveira R, C´ elio V, de Albuquerque N (2009). NECTAR:<br />
A DTN Routing Protocol Based on Neighborhood Contact History. SAC,<br />
pp. 8-12.<br />
Eyuphan B, Sabin CG, Boleslaw KS (2010).Conditional Shortest Path<br />
Routing in Delay Tolerant Networks. WoWMoM’, pp.1-6.<br />
Fall K, Farrell S (2008). DTN: an architectural retrospective. IEEE J.<br />
SelecT. Areas Commun., 5(26): 828-836.<br />
Guizhu W, Bingting W, Yongzhi G (2010). Dynamic Spray and Wait<br />
Routing algorithm with Quality of Node in Delay Tolerant Network.<br />
Jathar R, Gupta A (2010). Probabilistic Routing using Contact<br />
Sequencing in Delay Tolerant Networks.COMSNETS, pp. 1-10.<br />
Johnson DB, David A, Maltz J, Broch DSR (2001). The Dynamic Source<br />
Routing Protocol for Multi-Hop Wireless Ad Hoc Networks. Ad hoc.<br />
network. chapt., 5: 139-172.<br />
Lindgren A, Doria A, Schelen O (2003). Probabilistic routing in<br />
intermittently connected networks. SIGMOBILE Mobile Comput.<br />
Commun. Rev., 7(3).<br />
Qaisar A, Sulma RMSMZ (2011). TMHF: Transmit Max Hop First
1808 Sci. Res. Essays<br />
forwarding Strategy to Optimize the Performance of Epidemic<br />
Routing Protocol. Int. J. Comput. Appl., 18(5): 40-45.<br />
Srinivasa S, Krishnamurthi S (2009). CREAST: An opportunistic<br />
forwarding protocol based on conditional residual time. SECON’09.<br />
pp. 1-9.<br />
Thrasyvoulos S, Konstantinos P, Cauligi SR (2005). Spray and wait: an<br />
efficient routing scheme for intermittently connected mobile networks.<br />
WDTN,<br />
Thrasyvoulos S, Konstantinos P, Cauligi SR (2007). Spray and Focus:<br />
Efficient Mobility-Assisted Routing for Heterogeneous and Correlated<br />
Mobility. PERCOMW,<br />
Thrasyvoulos S, Konstantinos P, Cauligi SR (2008). Efficient routing in<br />
intermittently connected mobile networks: the multiple-copy case.<br />
IEEE/ACM Trans. Network., 16(1): 77-90.<br />
Vahdat A, Becker D (2000). Epidemic routing for partially connected ad<br />
hoc networks. Duke University, Tech. Rep.
Scientific Research and Essays Vol. 7(18), pp. 1809-1812, 16 May, 2012<br />
Available online at http://www.academicjournals.org/SRE<br />
DOI: 10.5897/SRE12.255<br />
ISSN 1992-2248 ©2012 <strong>Academic</strong> <strong>Journals</strong><br />
Full Length Research Paper<br />
Length-weight relationships, relative condition factor<br />
and relative weight of three fish species from beach<br />
seine fishing grounds in Iranian coastal waters of<br />
Caspian Sea<br />
Moradinasab, Gh. 1 *, Raeisi, H 1 , Paighambari, S.Y. 1 , Ghorbani, R 1 and Bibak, Z. 1<br />
Department of Fisheries, Faculty of Fisheries and Environment, Gorgan University of Agricultural Sciences and Natural<br />
Resources, Gorgan, Iran.<br />
Accepted 25 April, 2012<br />
The aim of this study is to record the length-weight relationship, relative condition factor (Krel) and<br />
relative weight (Wr) for three fish species in Iranian coastal waters of the Caspian Sea. Fish sampling<br />
was carried out in the beach seine fishing grounds in autumn and winter seasons for two years (2007<br />
and 2009). 14104 specimens were measured and weighed. The values of the exponent b in the lengthweight<br />
relationships (LWRs) were 2.8449 for Cyprinus carpio, 2.8844 for Liza aurata and 2.9077 for<br />
Rutilus frisii kutum. Relative condition factor (Krel) values were ranged from 1.017±0.002 to 1.071±0.002.<br />
In addition, Relative weight (Wr) ranged from 0.929±0.002 to 1.740±0.004.<br />
Key words: Length-weight relationship, relative condition factor, relative weight, Caspian Sea.<br />
INTRODUCTION<br />
The Caspian Sea is the largest lake in the world that Iran<br />
is associated with that via the coasts of Guilan,<br />
Mazandaran and Golestan provinces (Alizadeh, 2004).<br />
This lake has a rich diversity of aquatic species. One of<br />
the common methods of fishing in this lake is the beach<br />
seine fishing which accounts for about 60% of total<br />
fishing in the area (Iranian Fisheries Statistic Yearbook,<br />
2008). Rutilus frisii kutum (Kamensky, 1901), Liza aurata<br />
(Risso, 1810) and Cyprinus carpio (Linnaeus, 1758) are<br />
the most important fish species in fishing composition<br />
and they account for more than 95% of the beach seine<br />
fishing.<br />
In terms of economic, employment and income, the<br />
three species mentioned before play a significant role in<br />
people's lives and livelihoods (Abdolmalaki and<br />
Ghaninejad, 2005). R. frisii kutum is the most valuable<br />
species in this fishing method and it is widely reproduced<br />
*Corresponding author. E-mail: Moradinasab88@yahoo.com.<br />
Tel: +98 911 7098 500. Fax: +98 0171 2245886.<br />
artificially in reproduction and breeding farms in Iran, and<br />
release into the Caspian Sea every year (Razavi sayad,<br />
1999). This species lives only in the basin southern<br />
Caspian Sea and so far, it has not been reported its<br />
length-weight relationships in Fish base and neither the<br />
other two species (L. aurata and C. carpio) in Iran<br />
(Froese and Pauly, 2012).<br />
In this study, length-weight relationship (LWR), relative<br />
condition factor (Krel) and relative weight (Wr) for these<br />
three species is discussed.<br />
MATERIALS AND METHODS<br />
Data collection<br />
Fish specimens were collected monthly from 130 fishing grounds<br />
beach seine with a net mesh size of 28 to 33 mm STR (48˚ 52′ E,<br />
38˚ 19′ N to 53˚ 45′ E, 36˚ 25′ N, Iran) in autumn and winter<br />
seasons for two years (2007 and 2009). There was no sampling in,<br />
spring and summer because they are close seasons for beach<br />
seine fishing in Iran. Numbers of different sizes were separated of<br />
combination fishing randomly and then the fresh specimens<br />
measured and weighed using the total length (TL, nearest 0.1 cm)
1810 Sci. Res. Essays<br />
Table 1. Descriptive statistics and length-weight parameters for the three fish species in the fishing grounds beach seine Iranian coastal waters of Caspian Sea.<br />
Species n<br />
Length (cm) WLR parameters and statistics<br />
r 2<br />
Mean S.E Min Max a SE (a) 95% CL(a) b SE (b) 95%CL (b)<br />
Rutilus frisii kutum 4870 41.95 0.094 20.5 62.5 0.0179 0.022 0.0161-0.0198 2.9077 0.0139 2.8803-2.9350 0.91<br />
Cyprinus carpio 2090 46.08 0.181 22.7 72.4 0.0314 0.029 0.0275-0.0358 2.8449 0.0175 2.8104-2.8793 0.93<br />
Liza aurata 7144 36.55 0.075 21.3 60.5 0.0157 0.019 0.0144-0.0172 2.8844 0.0124 2.8599-2.9088 0.93<br />
n, number of individuals sampled; S.E, standard error; Min, minimum, Max, maximum; a, intercept; b, slope; CL 95%, confidence limits; r 2 , coefficient of determination.<br />
and body weight (BW, nearest 1 g), respectively. Total data<br />
were pooled together in each species without sexing.<br />
Data analysis<br />
The length-weight relationships were estimated by using<br />
following equation (Froese, 2006):<br />
W = a L b<br />
Where W is the body wet weight (g), L is the total length<br />
(cm), a is the intercept of the regression and b is the<br />
regression coefficient (slope). The parameters a and b of<br />
the length-weight relationships were estimated by the<br />
least-squares method based on logarithms:<br />
Log (W) = log (a) + b log (L)<br />
A t-test was used for comparison the b values obtained<br />
from the linear regressions with isometric values (Sokal<br />
and Rohlf, 1987):<br />
Where ts is the t-test value, b the slope and Sb the standard<br />
error of the slope (b). The comparison between the<br />
obtained values of t-test and the respective tabled critical<br />
values allowed the determination of the b values<br />
statistically significant, and their inclusion in an isometric<br />
(b=3) or allometric range (negative allometric; b3).<br />
In addition, for each individual, the relative condition<br />
factor (Krel) and the relative weight (Wr) were calculated by<br />
following the equations (Le cren, 1951; Froese, 2006;<br />
Wege and Anderson, 1978):<br />
Where ts is the t-test value, b the slope and Sb the standard<br />
error of the slope (b). The comparison between the<br />
obtained values of t-test and the respective tabled critical<br />
values allowed the determination of the b values<br />
statistically significant, and their inclusion in an isometric<br />
(b=3) or allometric range (negative allometric; b3).<br />
In addition, for each individual, the relative condition<br />
factor (Krel) and the relative weight (Wr) were calculated by<br />
following the equations (Le cren, 1951; Froese, 2006;<br />
Wege and Anderson, 1978):<br />
Where W is the body wet weight (g), L is the total length<br />
(cm), a and b are the parameters of length-weight<br />
relationships, Ws is a standard weight representing the75th<br />
percentile of observed weights at that length, am is<br />
geometric mean a and bm is geometric mean b. Statistics<br />
were performed using the R software version 2.11.0.<br />
RESULTS AND DISCUSSION<br />
A total of 14104 specimens of three fish species<br />
were collected from the fishing grounds beach<br />
seine Iranian coastal waters of Caspian Sea<br />
during the present study. The number of<br />
individuals sampled (n), the length and weight<br />
ranges, parameters a and b of the length-weight<br />
relationships, the standard error of b value and<br />
the determination coefficient (r 2 ) for the three<br />
species are given in Table 1.<br />
In this study, relative weight was obtained for R.<br />
frisii kutum 0.929±0.002. Several factors influence<br />
in the growth of the fish such as hereditary<br />
characteristics, food reserves, environmental<br />
factors, pollution, etc. Investigations carried out<br />
indicate that growth R. frisii kutum in the recent<br />
years has decreased (Abdolmalaki and<br />
Ghaninejad, 2005). This can be mainly attributed<br />
to artificial breeding and restocking programs<br />
carried out every year by the Iranian Fisheries<br />
Organization (Table 2).<br />
Over fishing and the elimination of larger<br />
individuals of this species due to the use of<br />
inappropriate fishing gears have also contributed<br />
to this situation. Considering that there is no<br />
selection involved in choosing male and female<br />
spawners for artificial breeding programs, the
Moradinasab et al. 1811<br />
Table 2. Relative condition factor (Krel) (±S.E) and Relative weight (Wr) (±S.E) for the three fish species in the fishing<br />
grounds beach seine Iranian coastal waters of Caspian Sea.<br />
Relative condition factor (Krel) Relative weight (Wr)<br />
Species Min Max Mean (S.E) Min Max Mean (S.E)<br />
Rutilus frisii kutum 0.43 2.57 1.017±0.002 0.38 2.32 0.929±0.002<br />
Cyprinus carpio 0.45 2.91 1.029±0.004 0.69 4.30 1.594±0.006<br />
Liza aurata 0.38 3.28 1.071±0.002 0.62 5.25 1.740±0.004<br />
Min, minimum; Max, maximum; S.E, standard error.<br />
Table 3. Length-weight relationships obtained from other parts of the world for L. aurata and C. carpio.<br />
Species Location and references<br />
Liza aurata<br />
Cyprinus carpio<br />
Type of<br />
length<br />
Length<br />
(cm)<br />
Sex a b<br />
Portugal; Algarve (Borges et al., 2003) TL 20.1 - 40.5 Unsexed 0.0078 3.006<br />
Croatia; Eastern Adriatic, (Dulcic and Glamuzina, 2006) TL 21.5 - 44.2 Unsexed 0.0181 2.952<br />
Greece; G. Saronikos, (Stergiou and Moutopoulos, 2001) SL 15.5 - 21.0 Mixed 0.0078 3.230<br />
Japan; Shioda Plain, Nagano Prefecture, (Carlander, 1969) TL 31.5 - 57.0 Unsexed 0.0060 3.210<br />
Turkey; Lake Iznik, Marmara, (Tarkan et al., 2006) TL 14.2 - 48.8 Unsexed 0.0250 2.830<br />
Spain; Araquil River, Navarra, (Miranda et al., 2006) TL 7.1 - 59.0 Mixed 0.0120 3.070<br />
gene bank of this species is gradually shifting.<br />
Thus providing the required conditions for natural<br />
spawning in rivers where this species migrates to and or<br />
the semi-natural breeding of R. frisii kutum in earthen<br />
ponds where the selection of breeders is allowed even if<br />
only of a small size is stressed more than before<br />
(Khanipour and Valipour, 2009).<br />
A result of present study isn’t in agreement with reports<br />
of Stergiou and Moutopoulos (2001) and Carlander<br />
(1969) and is in agreement with reports of Tarkan et al<br />
(2006) (Table 3).<br />
In studies of population dynamics high condition factor<br />
values shows of favorable environmental conditions<br />
(such as: habitat and prey availability) and low values<br />
indicate less favorable environmental conditions<br />
(Blackwell et al., 2000). Relative condition factor (Krel) is<br />
commonly factor for indicate the condition of fish species.<br />
In our study, Liza aurata (1.071±0.002) had best<br />
performance, while Krel value in Rutilus frisii kutum<br />
(1.017±0.002) was lowest across caught species (Table<br />
2).<br />
Table 3 indicates a and b parameters of weight-length<br />
relationships of selected species obtained from other<br />
parts of the world. Our results mostly agreed with the<br />
sturgeon species studies given in Table 3. The difference<br />
of a and b can be affected area, sex, season, degree of<br />
stomach fullness, gonad maturity, health, habitat,<br />
nutrition (Tesch, 1971).<br />
ACKNOWLEDGEMENTS<br />
We thank Iranian Fisheries Organization and Beach<br />
seine fishing cooperative societies for cooperation.<br />
REFERENCES<br />
Abdolmalaki Sh, Ghaninejad D (2005). Report on stock assessment and<br />
composition of the commercial bony fishes on the southern Caspian<br />
Sea. Fisheries Research Institute. Bandar Anzali, p. 132.<br />
Alizadeh H (2004). Introduction to the Features of the Caspian Sea.<br />
Norbakhsh publ., p. 120.<br />
Blackwell BG, Brown ML, Willis DW (2000). Relative Weight (Wr) Status<br />
and Current Use in Fisheries Assessment and Management. Rev.<br />
Fish. Sci., 8: 1-44.<br />
Borges TC, Olim S, Erzini K (2003). Weight-length relationship for fish<br />
species discarded in commercial fisheries of the Algarve (southern<br />
Portugal). J. Appl. Ichthyol., 19(6): 394-396.<br />
Carlander KD (1969). Handbook of freshwater fishery biology. Volume<br />
1. The Iowa State University Press. Ames. Iowa. DOI<br />
http://dx.doi.org/.<br />
Dulcic J, Glamuzina B (2006). Length-weight relationships for selected<br />
fish species from three eastern Adriatic estuarine systems (Croatia).<br />
J. Appl. Ichthyol., 22: 254-256.<br />
Froese R (2006). Cube law, condition factor and Length-Weight<br />
relationships: history, meta-analysis and recommendations. J. Appl.<br />
Ichthyol., 22: 241-253.<br />
Froese R, Pauly D (2012). Fish base. World Wide Web Electronic<br />
Publication. Available at http://www.fishbase.org. (accessed on 16<br />
February 2012).<br />
Khanipour AA, Valipour A (2009). Kutum jewel of the Caspian Sea.<br />
Iranian Fisheries Research Organization. Tehran. Iran, p. 97.<br />
Iranian Fisheries Statistic Yearbook (2008). Statistic catch fishes the<br />
basin southern Caspian Sea. Budget and code office- Statistic group<br />
and fishing developing study. Tehran Iran, p. 56.<br />
Le cren ED (1951). The length-weight relationships and seasonal cycle<br />
in gonad weight and condition in the perch (Perca fluviatilis). Anim.<br />
Ecol., 20: 201-219.<br />
Miranda E, Oscoz J, Leunda PM, Escala MC (2006). Weight-length<br />
relationships of cyprinid fishes of the Iberian Peninsula. J. Appl.<br />
Ichthyol., 22: 297-298.<br />
Razavi sayad B (1999). Introduction to the ecology of the Caspian Sea.
1812 Sci. Res. Essays<br />
Iranian Fisheries Research Organization (IFRO). p. 90.<br />
Sokal RR, Rolf FJ (1987). Introduction to Biostatistics. 2nd Edition.<br />
Freeman. New, York, p. 363.<br />
Stergiou KI, Moutopoulos DK (2001). A review of length-weight<br />
relationships of fishes from Greek marine waters. Naga ICLARM Q.<br />
24(1 and 2): 23-39.<br />
Tarkan AS, Gaygusuz O, Acipinar H, Gürsoy C, Ozulug M (2006).<br />
Length-weight relationships of fishes from the Marmara region (NW-<br />
Turkey). J. Appl. Ichthyol., 22: 271-273.<br />
Tesch FW (1971). Age and growth. In: Ricker, W. E. ed., Methods for<br />
Assessment of Fish Production in Freshwaters. Blackwell Scientific<br />
Publications. Oxford, pp. 98-100.<br />
Wege GJ, Anderson RO (1978). Relative weight (Wr): A new index of<br />
condition of largemouth bass. In: New approaches to management of<br />
small impoundments. G. Novinger J. Dillard (Eds). Am. Fish. Soc.<br />
Spec. Publ.. Bethesda. MD, 5: 79-91.
Scientific Research and Essays Vol. 7(18), pp. 1813-1829, 16 May, 2012<br />
Available online at http://www.academicjournals.org/SRE<br />
DOI: 10.5897/SRE11.1815<br />
ISSN 1992-2248 © 2012 <strong>Academic</strong> <strong>Journals</strong><br />
Full Length Research Paper<br />
Design and development of standalone DSP prototype<br />
for QT interval processing and monitoring<br />
Goh Chun Seng*, Sh-Hussain Salleh, J. M. Najeb, I. Kamarulafizam, Mahyar Hamedi and Alias Md Noor<br />
Centre for Biomedical Engineering (CBE), Transportation Research Alliance, Universiti Teknologi Malaysia, 813010<br />
Skudai, Johor, Malaysia.<br />
Accepted 30 March, 2012<br />
This paper describes the development of stand alone DSP hardware for QT interval monitoring and<br />
assessment. The QT interval has been known to be an important indicator prior to Myocardial Infarction<br />
(MI) and it is important to observe and monitor any changes in the period of the QT interval. The system<br />
consists of three units. The first unit includes floating point digital signal processor, TMS320VC33 which<br />
is selected for the development of signal processing with memory circuit. The following unit includes the<br />
development of signal monitoring displays circuit with universal serial bus (USB) interface. The third<br />
unit includes the development of analog front end (AFE) circuit and merged with signal conditioning<br />
circuit. This paper discusses the principle of system design and the system has the advantages to<br />
maximize utilization by allowing modification, reconfiguration, portable, cost effective and run in standalone<br />
operation. As well as eligibility of on board pre-program algorithm for QT analysis which existing<br />
ECG monitoring equipment are lacking. The system had been successfully tested with algorithm for QT<br />
analysis and it is validated with healthy subject’s data files from PTB diagnostic ECG data base.<br />
Key words: QT interval, DSP processor, stand alone operation, portable.<br />
INTRODUCTION<br />
The invention of ECG in the beginning of 20 th centuries<br />
by Willem Einthoven remains as demanded tool and<br />
popular component until today for non-invasive method to<br />
evaluate heart condition of patients who have been<br />
diagnose with signs and symptoms in clinical environment<br />
(Sandham et al., 2007). The interpretation of ECG<br />
will allow cardiologist and clinicians to understand and<br />
collect more information of electrical activity of the heart.<br />
However, with the integration of embedded technology,<br />
the automated interval measurement of P,Q,R,S and T in<br />
the ECG recording system has became an ideal<br />
instrument for patient monitoring and supervision in<br />
clinical practice. Although, the automated measurement<br />
are well establish in the ECG recording system but<br />
recently (Kligfield et al., 2006) there are studies focuses<br />
*Corresponding author. E-mail: seng4@yahoo.com. Tel: 07-<br />
5535933.<br />
on algorithm variation of automated QT interval<br />
measurement on commercial ECG recording system from<br />
different manufacturer.<br />
However, the automated methods of the QT interval<br />
are already widely available for computerized electrocardiography<br />
system (Christov, 2006), which ranges from<br />
manufactured ECG recording system to PC based<br />
system with different size and weight. Nevertheless,<br />
those commercial systems of automated QT interval<br />
measurements from different manufacturers have been<br />
utilize in studies to predict heart concern, (Kligfield et al.,<br />
2006) but due to different automated QT interval in the<br />
system with different automated algorithm from different<br />
manufacturers which can affect the identification and<br />
quantification of the acquired ECG, which means that<br />
different automated of QT measurement algorithm<br />
produces variation and error when validated against<br />
manually annotated QT database.<br />
The QT interval measurement is crucial and it is used<br />
as an informative index to predict sudden death due to
1814 Sci. Res. Essays<br />
cardiac arrhythmias (Malarvili et al., 2005). The indication<br />
of cardiac arrhythmias is related to the duration of the QT<br />
interval and it is important to observe and monitor any<br />
changes in the period of the QT interval (Mneimneh et al.,<br />
2006).<br />
Traditionally, QT interval measurements of patients are<br />
interpreted manually from ECG strip paper printed by<br />
ECG recording machines, the measurement are<br />
dependent and costly, if require skillful and experienced<br />
cardiologist or physician to carry out the manual<br />
procedures (Malik, 2004). Therefore the Food and Drug<br />
Administration (FDA) has initiated the encouragement to<br />
recommend digitizes ECGs and it is highlighted by<br />
Badilini (2005) as an efficient way to triggered the interest<br />
toward algorithm based ECG analysis.<br />
Given the fact that, commercial automated QT interval<br />
measurement with signal analysis capabilities is an<br />
expensive customize systems (Oweis and Hijazi, 2006),<br />
the reason are, the first; systems development timeline<br />
are strictly bounded by procedures from regulator.<br />
Second, it is intended to be use in clinical practice for<br />
human subject only (Saviola, 2005). Thus, only authorize<br />
manufacturer is given the mandate by regulator to make<br />
modifications and improvements toward improving speed<br />
and efficiency of the commercial system (Kaplan et al.,<br />
2004). Therefore, that restriction by regulator will be a<br />
financial challenge for researcher’s to acquire various<br />
types of instruments with QT interval measurement for<br />
just to harvest the raw ECG from patient and do offline<br />
QT analysis on computer. It is clearly to shows that, the<br />
commercial automated QT interval measurement<br />
systems are not subject for any research purposes in<br />
term of reprogrammable the QT algorithm on those<br />
commercial systems for ECG data analysis.<br />
Therefore, in this paper we introduce the development<br />
of an automated standalone system which is focused on<br />
portability, standalone operation and low cost QT interval<br />
measurement device. This standalone system is using<br />
Digital Signal Processor (DSPs) technology to<br />
accomplish the real time QT interval measurement from<br />
the tapped ECG and it is reprogrammable with simplest<br />
to more complex QT algorithms; ranges begin with simple<br />
beat to beat detection followed by advanced diagnosis for<br />
disorder heart disease. There are various systems<br />
applications in ECG parameters analysis have been<br />
reported (Mohammad et al., 2003; Gorup et al., 2000;<br />
Kara et al., 2006) but none of them describes a real time<br />
QT interval measurement regarding standalone operation,<br />
portability and low cost device.<br />
The high computational performance of Digital Signal<br />
Processor is sufficient to implement a standalone system<br />
without any host controller, such as computer. It is also a<br />
highly flexible standalone system and can be easily<br />
reprogrammed (Soon et al., 2001). Therefore, the<br />
advantages of the proposed standalone system, will allow<br />
cooperation between the researcher and cardiologist to<br />
work on implementation of QT algorithm development<br />
with on board reprogrammable function which is lacking<br />
in the existing commercial automated QT interval<br />
measurement systems.<br />
METHODOLOGY<br />
The architecture layout of the system<br />
The main components of the standalone system design will be<br />
described in stages according to the logically organization. The<br />
description will begin with Analog Front End Unit, follow by Signal<br />
Monitoring displays Unit, Signal Processing with Memory Unit and<br />
Power Supply Unit. All the stages of the process will be described.<br />
Figure 1 is the block diagram of the stand alone system and the<br />
way of components are interconnected and interrelated.<br />
Signal processing with memory unit<br />
This unit is focuses on incoming ECG from codec chip set for signal<br />
processing. There are few major components are used in the signal<br />
processing with memory unit. The mains components are 32 bit<br />
floating point DSP processor (TMS320VC33), Electrical Erasable<br />
Programmable Read only Memory - EEPROM (CAT28LV64W) and<br />
Static Random Access Memory - SRAM (CY7C1041DV33). The<br />
origin of DSP processor itself is design to handle mathematical and<br />
algorithm application and it is inefficient as a task oriented<br />
controller. For the SRAM memory in this unit is to allow for ECG<br />
data storage and the SRAM memory are expandable in both the<br />
data bus and address bus width.<br />
Analog Front End (AFE) Unit<br />
This unit provides interaction through peripherals connectivity with<br />
the real world environment. The peripherals are remote or<br />
pluggable analog front end devices either ready made or<br />
commercial or custom build ones to allow converted input signal to<br />
be delivered into the main board. Figure 1 shows that the tap ECG<br />
is delivered to the main board through the custom build AFE device.<br />
This AFE input consists of: (a) Signal conditioning circuitry (SCC),<br />
(b) Codec chip.<br />
The SCC is needed to amplify the tap ECG (0.5 to 5mV peak to<br />
peak) to at least the gain of 1000 (Najeb et al., 2005). Noises are<br />
common for tap ECG signal from body skin surface. Therefore, this<br />
SCC is build with instrumentation amplifier (INA121P) with low<br />
noise differential signal acquisition in order to minimize noise<br />
interference. The output from SCC is connected to codec chip set<br />
(PCM3003) to interface with TMS320VC33 in the main board.<br />
The codec chip is fully controlled by hardware setting and the<br />
data formats are 16 bit. Then the data transmission is only clock in<br />
and clock out in sequence from the codec chip set. So, it is need<br />
not depending on software setting or external clocking for<br />
operation.<br />
Signal monitoring unit<br />
This unit provides display on graphical liquid crystal displays<br />
(GLCD) for monitoring ECG as shown in Figure 1. In this unit, the<br />
microcontroller, PIC18LF452 is configured to interact with graphical<br />
liquid crystal displays (GLCD). The microcontroller is added in this<br />
unit without interrupting the TMS320VC33, because the<br />
microcontroller performs best in tasking order, which capable of
SCC<br />
Codec<br />
ADC<br />
Analog Front End unit<br />
Main Board<br />
Remote socket<br />
GLCD<br />
Microcontroller<br />
DSP<br />
RAM<br />
Signal monitoring<br />
displays unit<br />
Signal processing with<br />
Memory unit<br />
Seng et al. 1815<br />
USB<br />
ROM<br />
Figure 1. Block diagram of the stand alone system. The core processor is a floating point, Digital Signal Processor,<br />
TMS320VC33. The capture ECG signal is display on the Graphic LCD (128x64 dot pixel).l.<br />
task orientation and lack of DSP terms. However, the ADC0820<br />
received the input from the tap ECG signal through signal<br />
conditioning circuit (SCC) and the converted ADC data is latch to<br />
the output port of ADC0820. Then the tap ECG signal from SCC<br />
output was capture and converted by 8 bit Analog to Digital<br />
Converter chip set (ADC0820) and translated by PIC18LF452 into<br />
GLCD code in order to display on the GLCD.<br />
Figure 2, shows that, upon initialization, the microcontroller put to<br />
observe the FT245RL chip set to get acknowledge and the<br />
observation is done through Port A of PIC18LF452. As the<br />
acknowledge signal is detected, and then the firmware will proceed<br />
to next step’s label as ‘A’. If there is no feed back for acknowledge<br />
signal by the FT245RL, then in this case, the next stage will begin<br />
with monitoring the external interrupt and the firmware loops. If the<br />
interrupt is detected, PIC18LF452 will initialize the Graphic LCD<br />
panel and begin to read the ADC data from Port C of PIC18LF452.<br />
After that, the tap ECG is put on GLCD screen to display as the<br />
ECG monitoring signal.<br />
As depicted in Figure 3, there are two categories in the signal<br />
monitoring displays unit. The first category is the microcontroller<br />
with USB interfacing, colored with yellow dotted line. The FT245RL<br />
provides pin WR and pin RD# which are functions as write and read<br />
with 128 byte and 256 byte buffers in the FT245RL. The functions<br />
of the WR and the RD# pins are associated with pin TXE# and pin<br />
RXF# to indicate the buffer status, either it is empty so that written<br />
process is allowed to proceed or whether the buffer is full so that<br />
the read process can be executed. However, only pin WR and pin<br />
TXE# from FT245RL are utilized and both of these pins are<br />
connected to PIC18LF452. When TXE# is in LOW state means that<br />
the write buffer is ready and the data is awaiting at the FT245RL<br />
buffer. Then, the WR pin is toggled by PIC18LF452 from the HIGH<br />
state to LOW state, in order to transfer the data in bulks of 128<br />
bytes into the EEPROM. The EEPROM enable by setting both<br />
CE and WE pins are in LOW state and the OE pin is in HIGH<br />
stage.<br />
The lengths of the data in bulk mode are 21 bits and the transfer<br />
mode is in serial. Therefore, the 74HC164 with 8 parallel output is<br />
connected in cascade of 3 units in a row, so that this connection will<br />
allow the 21 bits of data to transfer as serial in and parallel out from<br />
the 74HC164. Figure 4 shows the circuit for microcontroller with<br />
USB interfacing and serial in parallel out circuit in signal monitoring<br />
displays unit. In Figure 5 shows the firmware loops and<br />
downloading process for pre-program ECG algorithm into the<br />
EEPROM in serial packet of address and data via USB connectivity.<br />
The firmware begins with the initialization of EEPROM and the<br />
shift register. After that, the download begins with sequence of 21<br />
bit per packet in serial format. Hence, the 21 bit per packet is<br />
shifting into the EEPROM by three shift register -74HC164 cascade<br />
in serial with each shift a byte of address and data into the<br />
EEPROM. The format of 21 bit per packet of address and data is<br />
depicted in Figure 6. The shift register is connected with FT245RL<br />
and it is monitored by PIC18LF452 to manage the elapse period of<br />
data transfer and will received acknowledge after data transaction<br />
is <strong>complete</strong>d.<br />
Power supply unit<br />
This prototype DSP hardware is designed to utilize the USB power.<br />
The focus is on low power consumptions. The USB voltage is<br />
regulated by TPS77533 and TPS77618 to provide necessary power<br />
demand in signal processing with memory unit, signal monitoring<br />
displays unit and analog front end (AFE) unit. Both of the voltage<br />
regulators are low drop out (LDP) voltage regulator. The ranges of<br />
different power supply are group as the following:<br />
i. 1.8V for DSP processor core supply.<br />
ii. 3.3V for DSP processor I/O pins and External Memory chip set.<br />
iii. 3.3V for Microcontroller.<br />
iv. 5.0V for Voltage regulator.
1816 Sci. Res. Essays<br />
QT Algorithm for ECG Signal Analysis<br />
Start<br />
Initialize:<br />
- Port A, Port B, Port C and Port D<br />
No<br />
FT245RL, get<br />
acknowledge?<br />
No<br />
Interrupt,<br />
ADC0820<br />
Yes<br />
?<br />
Initialize: - GCLD<br />
Yes<br />
Read the ADC data from Port C<br />
Display the digitize signal to GLCD as<br />
monitoring signal<br />
End<br />
Figure 2. Flowchart of firmware loop to display the tap ECG to the GLCD.<br />
The QT algorithm applied in this prototype DSP hardware is<br />
adapted from Laguna P et al. (1990). The criteria’s of these QT<br />
algorithm involves the preprocessing, QRS detection, R and Qwaves<br />
definition, QRS onset definition, T wave peak and T wave<br />
end definition. The criteria’s are depicted in Figure 7. The first<br />
attempt, a low pass differentiator (LDP) and first order low pass<br />
filter is used to eliminate noise (the residual and intrinsic<br />
differentiation noise). Next stage is the QRS detection, by defining<br />
threshold (H1) to detect the QRS peak. After detecting the QRS<br />
peak, the search of nearest peaks begins with forward and<br />
backward search, in order to identify the R peak. After that, the R<br />
and Q position (Rp and Qp) are define through zero crossing<br />
method. Then, continue with QRS onset definition, where<br />
backwards search is carry out in the differentiated signal d(k) for Qi<br />
and Ri in order to identify the present of Q and R wave (Laguna et<br />
al. 1990). Next stage, the QRS begin (QRS1) is define from the zero<br />
crossing point of Qi and Ri. From the R position Rp, a search<br />
A<br />
windows are define, the bwind and ewind. Then the searching is<br />
beginning with forward search to look for maximum and minimum<br />
value, which will define the T wave end in the searching process.<br />
This method is selected because it is produces the closet results<br />
(Moraes and Viana, 1997) obtained visually by specialist when<br />
tested for single lead (lead II). The QT algorithm was developed in<br />
C language with the assistant of Code Composer 3x/4x before it<br />
was implemented in the prototype DSP hardware. Upon Validation<br />
on this prototype DSP hardware for the developed QT algorithm, an<br />
annotated data set of Physikalisch Technische Bundesanstalt<br />
(PTB) Diagnostic ECG data prepared Christov et al. (2006) is used.<br />
As stated by Najeb (2007), the QT algorithm works best on healthy<br />
subjects in PTB diagnostic ECG data base. Therefore, this is done<br />
by downloading the PTB Diagnostic ECG healthy subject’s files<br />
only, from the data base into the prototype DSP hardware.<br />
The implementation and validation of this QT algorithm to the<br />
prototype DSP hardware is described in the flow diagram as<br />
depicted in Figure 8. The experiment setup procedure is to validate<br />
the QT algorithm for the prototype DSP hardware. The process
Seng et al. 1817<br />
Data /<br />
Address<br />
USB<br />
FT245RL<br />
EEPROM Shift Register<br />
Microcontroller<br />
TXE#<br />
WR<br />
CLK<br />
Data<br />
/WE<br />
/OE<br />
S.I.P.O<br />
Signal Monitoring Displays unit<br />
Figure 3. Block diagram of hardware interfacing in signal monitoring displays unit.<br />
CLR<br />
9<br />
CLK<br />
8<br />
A<br />
1<br />
B<br />
2<br />
QA<br />
3<br />
QB<br />
4<br />
QC<br />
5<br />
QD<br />
6<br />
QE<br />
10<br />
QF<br />
11<br />
QG<br />
12<br />
QH<br />
13<br />
VCC<br />
14<br />
GND<br />
7<br />
srU1<br />
CLR<br />
9<br />
CLK<br />
8<br />
A<br />
1<br />
B<br />
2<br />
QA<br />
3<br />
QB<br />
4<br />
QC<br />
5<br />
QD<br />
6<br />
QE<br />
10<br />
QF<br />
11<br />
QG<br />
12<br />
QH<br />
13<br />
VCC<br />
14<br />
GND<br />
7<br />
srU3<br />
srVCC<br />
srVCC<br />
srVCC<br />
srVCC<br />
srVCC<br />
srVCC<br />
srCLK_IN<br />
srCLK_IN<br />
srCLK_IN<br />
sr3D8<br />
sr2D8<br />
sr1D8<br />
sr1 D 8<br />
sr2 D 8<br />
CLR<br />
9<br />
CLK<br />
8<br />
A<br />
1<br />
B<br />
2<br />
QA<br />
3<br />
QB<br />
4<br />
QC<br />
5<br />
QD<br />
6<br />
QE<br />
10<br />
QF<br />
11<br />
QG<br />
12<br />
QH<br />
13<br />
VCC<br />
14<br />
GND<br />
7<br />
srU2<br />
1 2<br />
3 4<br />
5 6<br />
7 8<br />
9 10<br />
srJP1<br />
srVCC<br />
srD ata_ IN<br />
srCL K _ IN<br />
srD ata_ IN<br />
sr/W E<br />
sr/CE<br />
sr/O E<br />
A0<br />
10<br />
A1<br />
9<br />
A2<br />
8<br />
A3<br />
7<br />
A4<br />
6<br />
A5<br />
5<br />
A6<br />
4<br />
A7<br />
3<br />
A8<br />
25<br />
A9<br />
24<br />
A10<br />
21<br />
A11<br />
23<br />
A12<br />
2<br />
CE<br />
20<br />
OE<br />
22<br />
WE<br />
27<br />
GND<br />
14<br />
VCC<br />
28<br />
DQ0<br />
11<br />
DQ1<br />
12<br />
DQ2<br />
13<br />
DQ3<br />
15<br />
DQ4<br />
16<br />
DQ5<br />
17<br />
DQ6<br />
18<br />
DQ7<br />
19<br />
NC<br />
1<br />
NC<br />
26<br />
srU5<br />
sr/WE<br />
sr/CE<br />
sr/OE<br />
srVCC_S<br />
sr1D1<br />
sr1D2<br />
sr1D3<br />
sr1D4<br />
sr1D5<br />
sr1D6<br />
sr1D7<br />
sr1D8<br />
sr1D1<br />
sr1D2<br />
sr1D3<br />
sr1D4<br />
sr1D5<br />
sr1D6<br />
sr1D7<br />
sr2D1<br />
sr2D2<br />
sr2D3<br />
sr2D4<br />
sr2D5<br />
sr2D6<br />
sr2D7<br />
sr3D1<br />
sr3D2<br />
sr3D3<br />
sr3D4<br />
sr3D5<br />
sr3D6<br />
sr3D7<br />
sr2D8<br />
sr2D1<br />
sr2D2<br />
sr2D3<br />
sr2D4<br />
sr2D5<br />
sr2D6<br />
sr2D7<br />
sr3D1<br />
sr3D2<br />
sr3D3<br />
sr3D4<br />
sr3D5<br />
1 2<br />
3 4<br />
5 6<br />
srJP2<br />
srVCC_S<br />
srVCC_S<br />
w1 w1 w3<br />
w3<br />
vr5V ldo3V<br />
1<br />
2<br />
uY1 20Mhz<br />
uC7<br />
22p<br />
uC8<br />
22p<br />
uS1<br />
10K<br />
uR13<br />
uVP5<br />
470<br />
uR12<br />
RC7/RX/DT<br />
26<br />
RD4/PSP4<br />
27<br />
RD5/PSP5<br />
28<br />
RD6/PSP6<br />
29<br />
RD7/PSP7<br />
30<br />
VSS<br />
31<br />
VDD<br />
32<br />
RB0/INT0<br />
33<br />
RB1/INT1<br />
34<br />
RB2/INT2<br />
35<br />
RB3/CCP2*<br />
36<br />
RB4<br />
37<br />
RB5/PGM<br />
38<br />
RB6/PGC<br />
39<br />
RB7/PGD<br />
40<br />
RC6/TX/CK<br />
25<br />
RC5/SDO<br />
24<br />
RC4/SDI/SDA<br />
23<br />
RD3/PSP3<br />
22<br />
RD2/PSP2<br />
21<br />
RE2/^CS/AN7<br />
10<br />
RD1/PSP1<br />
20<br />
RD0/PSP0<br />
19<br />
RC3/SCK/SCL<br />
18<br />
RC2/CCP1<br />
17<br />
RC1/T1OSL/CCP2*<br />
16<br />
RC0/T1OSO/T1CKL<br />
15<br />
OSC2/CLKO/RA6<br />
14<br />
OSC1/CLKL<br />
13<br />
VSS<br />
12<br />
VDD<br />
11<br />
RE1/^WR/AN6<br />
9<br />
RE0/^RD/AN5<br />
8<br />
RA5/AN4/^SS/LVDIN<br />
7<br />
RA4/T0CKL<br />
6<br />
RA3/AN3/VREF+<br />
5<br />
RA2/AN2/VREF-<br />
4<br />
RA1/AN1<br />
3<br />
RA0/AN0<br />
2<br />
/MCLR/VPP<br />
1<br />
uU2<br />
1<br />
2<br />
3<br />
4<br />
5<br />
6<br />
7<br />
8<br />
9<br />
10<br />
11<br />
12<br />
13<br />
14<br />
15<br />
16<br />
17<br />
18<br />
19<br />
20<br />
21<br />
22<br />
23<br />
24<br />
25<br />
26<br />
27<br />
28<br />
29<br />
30<br />
31<br />
32<br />
33<br />
34<br />
35<br />
36<br />
37<br />
38<br />
39<br />
40<br />
uJP5<br />
1<br />
2<br />
3<br />
4<br />
5<br />
6<br />
7<br />
8<br />
9<br />
10<br />
11<br />
12<br />
13<br />
14<br />
15<br />
16<br />
17<br />
18<br />
19<br />
20<br />
21<br />
22<br />
23<br />
24<br />
25<br />
26<br />
27<br />
28<br />
29<br />
30<br />
31<br />
32<br />
33<br />
34<br />
35<br />
36<br />
37<br />
38<br />
39<br />
40<br />
uJP4<br />
uVP5<br />
uVP5<br />
VBUS<br />
1<br />
D-<br />
2<br />
D+<br />
3<br />
GND<br />
4<br />
uUSB_B1<br />
1<br />
2<br />
uJP3<br />
uVP5<br />
USBDM<br />
USBDP<br />
4K7<br />
uR5<br />
10K<br />
uR6<br />
4.7uF<br />
uC5<br />
104pF<br />
uC3<br />
104pF<br />
uC2<br />
uVP5<br />
104pF<br />
uC1<br />
uLD8<br />
uLD9<br />
uVP5<br />
1<br />
2<br />
3<br />
4<br />
5<br />
6<br />
7<br />
8<br />
9<br />
10<br />
11<br />
12<br />
13<br />
14<br />
15<br />
16<br />
17<br />
18<br />
19<br />
20<br />
21<br />
22<br />
23<br />
24<br />
25<br />
26<br />
27<br />
28<br />
uJP2<br />
D0<br />
1<br />
D4<br />
2<br />
D2<br />
3<br />
VCCIO<br />
4<br />
D1<br />
5<br />
D7<br />
6<br />
GND<br />
7<br />
NC<br />
8<br />
D5<br />
9<br />
D6<br />
10<br />
D3<br />
11<br />
PWREN#<br />
12<br />
RD#<br />
13<br />
WR<br />
14<br />
USBDP<br />
15<br />
USBDM<br />
16<br />
3V3OUT<br />
17<br />
GND<br />
18<br />
RESET#<br />
19<br />
VCC<br />
20<br />
GND<br />
21<br />
TXE#<br />
23<br />
RXF#<br />
22<br />
NC<br />
24<br />
AGND<br />
25<br />
TEST<br />
26<br />
OSCI<br />
27<br />
OSCO<br />
28<br />
uU1<br />
1<br />
2<br />
3<br />
4<br />
5<br />
6<br />
7<br />
8<br />
9<br />
10<br />
11<br />
12<br />
13<br />
14<br />
15<br />
16<br />
17<br />
18<br />
19<br />
20<br />
21<br />
22<br />
23<br />
24<br />
25<br />
26<br />
27<br />
28<br />
uJP1<br />
USBDM<br />
USBDP<br />
Figure 4. Serial in Parallel out (SIPO) circuit and EEPROM interface circuit.
1818 Sci. Res. Essays<br />
A<br />
Initialize:<br />
- Shift Register and EEPROM<br />
Data Packet:<br />
- Address byte and Byte count<br />
Start <strong>Download</strong>ing to EEPROM<br />
Done?<br />
End<br />
Yes<br />
Figure 5. Flow chart of downloading data to EEPROM.<br />
Location 1<br />
Location 2<br />
Location 3<br />
1 st Address<br />
2 nd Address<br />
3 rd Address<br />
+<br />
+<br />
+<br />
21 bit<br />
No<br />
1 st Data<br />
2 nd Data<br />
3 rd Data<br />
Location n n th Address n th + Data<br />
Figure 6. Transfer data packet of 21 bit to EEPROM via USB.<br />
begins with hardware setups followed by Code Composer to load<br />
the QT algorithm from the created workspace window and deliver<br />
each file (healthy subjects) from the PTB Diagnostic ECG data base<br />
by using probe point into the prototype DSP hardware. The data<br />
file format for PTB Diagnostic ECG data base is converted into the<br />
Code Composer 3x/4x format. As the processing begins, the signal<br />
transition is visible on the graphical display window as shown in<br />
Figure 9 for the DSP prototype hardware so that it enables<br />
researcher to observe and gain insight for each executed<br />
processing step. Finally, the obtained result for Q onset and T offset<br />
are displayed at the watch window for the prototype DSP hardware<br />
as depicted in Figure 9. The obtained results are validated with the<br />
annotated data set.<br />
The ECG data for validation uses in Code Composer 3x/4x is<br />
acquired from PTB diagnostic ECG database. In this ECG database<br />
only information of ECG lead II is extracted for QT interval<br />
measurement. However, the extracted lead II information are in<br />
numeric text format but the required format for File I/O using probe<br />
point in Code Composer 3x/4x, must be in hexadecimal. Therefore,<br />
conversion of file format is required and needed to be done in order<br />
for Code Composer 3x/4x to read those ECG files to the target<br />
processor, TMS320VC33.<br />
The requirements for the requested ECG files are in 2 bytes<br />
format for each data present in the converted file from the PTB<br />
Diagnostic ECG data base.<br />
Figure 10 shows a converted file of patient104_s0306lre taken<br />
from PTB Diagnostic ECG data base: On the left of Figure 10<br />
shows the raw PTB Diagnostic ECG data with file name:
Preprocessing<br />
- Low pass differentiator (LPD)<br />
- 1 st order low pass filter.<br />
QRS detection<br />
- Define Threshold (H1):QRSmax and QRS min<br />
- Identified R peak.<br />
R and Q waves definition<br />
- Define R and Q position (Rp and Qp) from<br />
zero crossing.<br />
QRS onset definition<br />
- QRS-begin is define from the zero<br />
crossing of Rp and Qp.<br />
T-wave peak and T-wave end definition<br />
- Define a search window from the Rp.<br />
- bwind and ewind, forward search.<br />
Figure 7. Simplified flow chart for Q onset and T offset<br />
detection.<br />
Start<br />
Setup DSP Hardware<br />
Initialize Code Composer<br />
Load workspace<br />
Load Program (QT algorithm)<br />
Compile and build<br />
Load “QT algorithm’’ onto DSP processor<br />
Set and Connect Probe Point to acquired downloaded files’ from PTB<br />
Diagnostic ECG database (Healthy subjects)<br />
Observe signal transition (Graphical Display Window)<br />
Display Q Onset and T Offset results (Watch Windows)<br />
Validate the results with annotated Q onset and T offset data set<br />
End<br />
Figure 8. Flow diagram of QT algorithm implementation and validation for the developed prototype DSP<br />
hardware.<br />
Seng et al. 1819
1820 Sci. Res. Essays<br />
ECG file<br />
Watch window: Measured result for QT interval<br />
Signal transition<br />
Figure 9. The QT algorithm is developed in C with the assistant of Code Composer 3x/4x. The signal transition is<br />
visualized at the watch windows for each preprocessing interface to the prototype DSP hardware.<br />
Raw<br />
ECG Data, from PTB<br />
Diagnostic ECG<br />
data base<br />
Converted<br />
ECG data<br />
File Header<br />
Figure 10. The ECG file conversion from raw ECG data to 2 byte format, to be used with Code Composer 3x/4x.
Fix number<br />
Format Starting address Page number<br />
Samples data in the file<br />
Length<br />
Figure 11. File header information, the converted ECG file (patient104_s0306lre) with 2 byte format.<br />
patient104_s0306lre is extracted from Physio bank ATM for lead II<br />
information and on the right is the converted information, in 16 bit or<br />
2 byte format. The converted ECG file of patient104_s0306lre<br />
begins with a header format located in the first line of the files.<br />
The header format is depicted in Figure 11. The first line of the<br />
converted file is the header’s information which is in hexadecimal<br />
format. The header consists of useful and needed information for<br />
ECG file processing. The header information syntax is a) fix<br />
number, b) data format, c) starting address of memory block, d)<br />
page number e) length of block data.<br />
(a) Fix number: Fixed by default numeric number, 1651.<br />
(b) Data format: Is a numeric number beginning with 1 to 4, to<br />
indicate the samples format in the file. This numeric number<br />
represents a data format: hexadecimal, integer, long integer,<br />
floating point, respectively. However, numeric 1 is selected to<br />
indicate that the entire converted ECG file for patient104_s0306lre<br />
is in hexadecimal.<br />
(c) Starting address: To indicate the start address for data file. In<br />
this case the setting is 0.<br />
(d) Page number: Is a numeric number beginning with 1 to 2, to<br />
indicate the type of file contents. This numeric number represents:<br />
data and program respectively. However, numeric 1 is selected to<br />
indicate the file contents are only data.<br />
(e) Length: To indicate the total samples data in the file. In this<br />
case the setting is 0.<br />
Nevertheless, there are two numeric zero is defined in the file<br />
header as shown in Figure 11. These two numeric zeros are<br />
representing the starting address of data file and the length of data<br />
Seng et al. 1821<br />
file. But, these two zeros can be overwritten, when using the Code<br />
Composer data file format with file I/O capabilities as depicted in<br />
Figure 12, this means that any information entered in the Address<br />
dialog box and in the Length dialog box, will automatically override<br />
the Code Composer data file information. Followed by, the<br />
information which is entered, to the dialog box, the address begins<br />
with inp_buffer and the lengths are 3200 samples of data. The 3200<br />
samples of data are the converted ECG file of patient104_s0306lre<br />
as shown in Figure 10. The address and the length of data also<br />
defined in C, respectively as depicted in Figure 12b. In order to link<br />
the file, as shows in Figure 12c it is needed to connect probe point<br />
with the converted ECG file of patient104_s0306lre. After that, is to<br />
execute the QT algorithm under Code Composer 3x/4x as depicted<br />
in Figure 12d.<br />
RESULTS<br />
Experiment had been carried out in order to determine<br />
the functionality and performance for this standalone<br />
DSP prototype design. The ECG parameter chosen for<br />
this experiment setup were Q onset and T offset.<br />
Therefore, this prototype was validated by using the PTB<br />
diagnostic ECG database utilizing the annotated QT data<br />
set prepared by Christov et al. (2006). An emulator test<br />
setup with code composer 3x/4x were used to test the<br />
standalone prototype DSP hardware functionally and
1822 Sci. Res. Essays<br />
a) Setup address and length<br />
for input buffer<br />
Correction<br />
b) Address and buffer for the data length using<br />
the C file system in Code Composer 3x/4x<br />
c) A converted ECG file of patient104_s0306lre is connected with probe point.<br />
d) The QT algorithm read in 3200 word of data from a converted ECG file of patient104_s0306lre.<br />
Figure 12: 12. Screen snapshoot of of linking a file a file with with Code Code Composer 3x/4x 3x/4x to the to prototype the prototype DSP hardware. DSP hardware.<br />
performance.<br />
The test setup was organized into two parts: a) the<br />
functional test results and b) the validation result. In<br />
functional test results, tap ECG were acquired from AFE<br />
unit to be displayed on the graphical LCD (GLCD) display<br />
and the on chip (TMS320VC33) physical pin H1 is<br />
checked to ensure the present of signal oscillation in<br />
multiple of scale five factors from the clock input<br />
(oscillator). For validation results, ECG file from PTB<br />
diagnostic ECG database were used and compared with<br />
the annotated data set, the obtained results were plot in<br />
line graph. The functional test and validation results will<br />
be discussed.<br />
Functional test results<br />
When power on, pin H1 was checked. The result is
100MHz<br />
20MHz<br />
Figure 13. Pin H1 at TMS320VC33, capture by scope-meter software (fluke, model-199b).<br />
shown in Figure 13, the 100MHz is the multiple x5 scale<br />
factor from the clock input which is the oscillator<br />
frequency, 20MHz. This proves that the chip set<br />
(TMS320VC33) is operational. After that, ECG signal is<br />
acquired through the input from AFE as shown Figure 14.<br />
The AFE hardware is custom build and the ECG signal is<br />
obtain from the Lead II formation on human body skin<br />
surface with fours disposable electrodes place on the<br />
right hand, left hand, right leg and left leg.<br />
This custom build AFE hardware device consists of: a)<br />
Signal Conditioning Circuitry (SCC) and b) codec chip,<br />
PCM3003. The SCC is needed to amplify the tap ECG<br />
(0.5mV to 5mV peak to peak) to at least the gain of 1000<br />
(Najeb et al., 2005). Noises are common for tap ECG<br />
signal from body skin surface. Therefore, this SCC is<br />
build with instrumentation amplifier (INA121P) with low<br />
noise differential signal acquisition in order to minimize<br />
noise interference. The tap ECG is connected to scopemeter<br />
(fluke, model-199b) at SCC output, as shown in<br />
Figure 14(a). This shows that, the SCC is operational.<br />
Then the tap ECG signal from SCC output was capture<br />
and converted by 8 bit Analog to Digital Converter chip<br />
set (ADC0820) and translated by PIC18LF452 into GLCD<br />
code. As a result, the tap ECG signals are able to display<br />
at the GLCD as shown in Figure 14(b).<br />
Seng et al. 1823<br />
The output from SCC is connected to codec chip set to<br />
interface with TMS320VC33 in the main board. PCM3003<br />
is selected because it is fully controlled by hardware<br />
setting and the data formats are selectable (16 bit or 20<br />
bit) by physical pins configuration, but in this work the<br />
data format 16 bit is selected. Then the data transmission<br />
is only clock in and clock out in sequence from the codec<br />
chip set. So, it is not dependent on software setting or<br />
external clocking for operation.<br />
The connected circuitry depicted in Figure 14 shows<br />
that the ECG monitoring signal is directly send to<br />
microcontroller without interrupting the TMS320VC33.<br />
This means that, the TMS320VC33 is only focus on<br />
incoming ECG from codec chip set for signal processing.<br />
Figure 15 depicted two different tap ECG monitoring<br />
displays result for test subject resting in static and non<br />
static condition.<br />
Validation result: Q Onset and T Offset<br />
The validation of the prototype DSP hardware performance<br />
for QT algorithm is carried out on the PTB<br />
diagnostic ECG database by utilizing the annotations<br />
prepared by Christov et al. (2006). Table 1 shows the
1824 Sci. Res. Essays<br />
a b<br />
SCC<br />
Codec<br />
ADC<br />
Analog Front End unit<br />
Main Board<br />
Remote socket<br />
GLCD<br />
Microcontroller<br />
DSP<br />
RAM<br />
Signal monitoring<br />
displays unit<br />
USB<br />
ROM<br />
Signal processing<br />
with<br />
Memory unit<br />
Figure 14. The tap Electrocardiogram capture by scope-meter (fluke, model-199b) and display as monitoring signal<br />
on Graphic LCD.<br />
selected 15 files for healthy control subjects with<br />
annotated Q onset (in millisecond - ms) used in validating<br />
the prototype DSP hardware. The obtained deviations<br />
mean errors for prototype DSP hardware is 9.13 ms.<br />
The prototype design has been tested with algorithm<br />
for QT analysis and it is validated with 15 healthy<br />
subject’s files from PTB diagnostic ECG data base. The<br />
recording length of each selected file by default is about<br />
one minute trace. Each selected test file data contain one<br />
selected beat marked visually by referee’s annotation<br />
(Christov et al., 2006). Table 1 shows that, there are<br />
selected 15 healthy subject files from the PTB Diagnostic<br />
ECG database are used to obtain the measured results<br />
from prototype DSP hardware. The measured results of<br />
the prototype DSP hardware were presented in line graph<br />
as depicted in Figure 16. The yellow illustrates the<br />
annotated Q onset and the purple illustrates the prototype<br />
DSP hardware. The overall measured results trend in<br />
Figure 16 shows fluctuation with slight increased towards<br />
the end of the files listing. However, the comparison of<br />
the measured results with the annotated Q onset shows<br />
moderate differences. There are five patient file shows<br />
higher deviation, which is more than 10 ms and above.<br />
Those patient files are patient166/s0275lre, patient172/<br />
s0304lre, patient180/s0561_re, patient198/s0402lre and<br />
patient263/s0499_re.<br />
This is the second part of validation for the prototype<br />
DSP hardware, which is validate on T offset detection.<br />
The total of 15 patient files was acquired from PTB<br />
diagnostic ECG database, T offset detection for each of
(a) Test Subject: Resting in static<br />
condition<br />
(b) Test Subject: Resting in non-static condition<br />
Figure 15: ECG based line moment due to test subject body movement. (a)<br />
In static condition. (b) In non-static condition.<br />
Table 1. Comparison of Q onset between the annotated data set with the measured<br />
one and the obtained deviation results for prototype DSP hardware.<br />
PTB Diagnostic ECG Data Base Prototype DSP Hardware<br />
No Patient File<br />
Q onset (ms)<br />
Annotated<br />
Q onset (ms)<br />
Measured Deviation<br />
1 patient104/s0306lre 750 756 6<br />
2 patient116/s0302lre 1136 1140 4<br />
3 patient165/s0322lre 1221 1224 3<br />
4 patient166/s0275lre 1632 1612 20<br />
5 patient172/s0304lre 1470 1486 16<br />
6 patient180/s0561_re 1751 1732 19<br />
7 patient198/s0402lre 1165 1192 27<br />
8 patient233/s0457_re 1543 1540 3<br />
9 patient234/s0460_re 1518 1520 2<br />
10 patient241/s0469_re 1717 1716 1<br />
11 patient241/s0470_re 1566 1568 2<br />
12 patient243/s0472_re 1155 1152 3<br />
13 patient248/s0481_re 1061 1056 5<br />
14 patient263/s0499_re 1473 1496 23<br />
15 patient264/s0500_re 1389 1392 3<br />
Mean error 9.13<br />
Seng et al. 1825
1826 Sci. Res. Essays<br />
millisecond (ms)<br />
2000<br />
1800<br />
1600<br />
1400<br />
1200<br />
1000<br />
800<br />
600<br />
400<br />
200<br />
0<br />
750<br />
756<br />
1136<br />
1140<br />
1632<br />
1612<br />
1221<br />
1224<br />
1470<br />
1486<br />
Annotated Data set Prototype DSP Hardware<br />
1751<br />
1732<br />
1165<br />
1192<br />
1543 1518<br />
1540 1520<br />
1717 1566<br />
1716 1568<br />
1155 1061<br />
1152 1056<br />
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15<br />
Patient File<br />
Figure 16. Comparison of annotated (yellow) and the measured result, Q onset in ms (millisecond). Purple illustrates the measured Q<br />
onset on prototype DSP hardware.<br />
Table 2. Comparison of T Offset between the annotated data set with the measured one and the<br />
obtained deviation results for prototype DSP hardware.<br />
PTB Diagnostic ECG Data Base Prototype DSP Hardware<br />
No Patient File<br />
T Offset (ms)<br />
Annotated<br />
T Offset (ms)<br />
Measured Deviation<br />
1 patient104/s0306lre 1136 1132 4<br />
2 patient116/s0302lre 1540 1525 15<br />
3 patient165/s0322lre 1612 1612 0<br />
4 patient166/s0275lre 2043 2036 7<br />
5 patient172/s0304lre 1878 1876 2<br />
6 patient180/s0561_re 2123 2122 1<br />
7 patient198/s0402lre 1530 1528 2<br />
8 patient233/s0457_re 1939 1932 7<br />
9 patient234/s0460_re 1950 1948 2<br />
10 patient241/s0469_re 2114 2112 2<br />
11 patient241/s0470_re 1962 1960 2<br />
12 patient243/s0472_re 1476 1468 8<br />
13 patient248/s0481_re 1448 1440 8<br />
14 patient263/s0499_re 1829 1818 11<br />
15 patient264/s0500_re 1797 1794 3<br />
Mean error 4.93<br />
the file was annotated with a marker by utilizing the<br />
annotations prepared by Christov et al. (2006). Table 2<br />
depicted the 15 healthy control subjects files with<br />
annotated T offset (in millisecond - ms) used for<br />
validating the prototype DSP hardware. The measured T<br />
offset results for the deviation mean errors are 4.93 ms.<br />
The measured results of 15 healthy subject files (PTB<br />
Diagnostic ECG database) from Table 2 for the prototype<br />
1473<br />
1496<br />
1389<br />
1392<br />
DSP hardware were presented in line graph as depicted<br />
in Figure 17. The blue illustrates the annotated T offset<br />
and the purple illustrates the prototype DSP hardware. As<br />
an overall trend, it is clear that, there were wild<br />
fluctuations in T offset measured results towards the end<br />
of the files listing in Figure 17 which represented the<br />
prototype DSP hardware. However, the measured results<br />
of the prototype DSP hardware are close to the
millisecond(ms)<br />
2500<br />
2000<br />
1500<br />
1000<br />
500<br />
0<br />
1136<br />
1132<br />
1540<br />
1525<br />
1612<br />
1612<br />
2043<br />
2036<br />
Annotated Data set Prototype DSP Hardware<br />
1878<br />
1876<br />
1950<br />
2123 1948<br />
2122<br />
1939<br />
1932<br />
1530<br />
1528<br />
2114<br />
2112<br />
1962<br />
1960<br />
1476<br />
1468<br />
1448<br />
1440<br />
Seng et al. 1827<br />
1829<br />
1818<br />
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15<br />
Patient File<br />
1797<br />
1794<br />
Figure 17. Comparison of annotated (blue) and the measured result, T offset in ms (millisecond). Purple illustrates the<br />
measured T offset on prototype DSP hardware.<br />
annotated T offset. It can be clearly seen that, the highest<br />
deviation of measured results is the second file,<br />
patient116/s0302lre which is 15 ms, and the lowest or no<br />
deviation is the third file, patient165/s0322lre which is 0<br />
ms.<br />
DISCUSSION<br />
The prototype DSP hardware has been built, this<br />
prototype is validated using PTB ECG diagnostic<br />
database and the acquired tap ECG is able to put on<br />
Graphic LCD. In summary, the <strong>complete</strong>d prototype DSP<br />
hardware system is depicted in Figure 18. The Analog<br />
Front End (AFE) unit is interfaced with volunteer subject<br />
as shown in Figure 18(b) and Figure 18(d). The output of<br />
AFE is seen at the oscilloscope screen in Figure 18(e),<br />
this shows that it is operational. The signal processing<br />
with memory unit is shown in Figure 18(a). However, in<br />
Figure 18(c) is the signal monitoring displays unit to<br />
display the output of AFE on GLCD with various ECG<br />
based line moment pattern either the volunteer subject is<br />
resting in static condition or non static condition, the<br />
display results are shown in Figure 18(f) and Figure<br />
18(g). The validation process on the prototype DSP<br />
hardware for QT algorithm is able to be displayed on the<br />
computer screen as shown in Figure 18(h) through the<br />
emulator.<br />
Nevertheless, the ECG data for this prototype DSP<br />
hardware validation is taken directly from PTB diagnostic<br />
ECG data base therefore is not influenced by the codec.<br />
Codec is introduced initially in the system as an initial<br />
development for this system. The codec will be replace<br />
with ADC in the future work for real time QT<br />
measurement on standalone DSP platform. Furthermore,<br />
the functional test as depicted earlier shows that this<br />
prototype is operational. Since with the proposed of this<br />
prototype DSP hardware is targeted at bridging the gap<br />
among researcher and cardiologist have for in<br />
collaboration to develop and implementation of QT<br />
algorithm at a low cost standalone DSP platform. As well<br />
as eligibility of on board pre-program algorithm for QT<br />
analysis which existing ECG monitoring equipment are<br />
lacking. For the presented prototype DSP hardware<br />
design can be extended in term of design, method and<br />
structure to capture and conduct other types of<br />
biomedical signal processing in real time processing.<br />
Conclusions<br />
In this paper, we present a design and development of<br />
standalone DSP prototype for QT Interval processing and<br />
monitoring. The high cost and lack of flexible applications<br />
on existing DSP hardware in the market has prompted<br />
researcher to design and fabricate a low cost with applications<br />
versatility DSP hardware. By the implementation<br />
of an operational standalone prototype DSP hardware<br />
design, the stand alone prototype DSP hardware system<br />
using floating point digital signal processor has been<br />
successfully built. The building of the system begins by<br />
fabricating the main board as the core processing unit<br />
and the analog front end device as the input. The core<br />
unit for main board signals processing is designed with a<br />
Digital Signal Processor (DSPs), TMS320VC33. The QT<br />
algorithm is applied in this prototype DSP hardware,<br />
whereby it is validated with PTB diagnostic ECG data<br />
base and the mean error results are; Q onset is 9.13 ms
1828 Sci. Res. Essays<br />
h<br />
d<br />
Prototype DSP hardware ECG Monitor<br />
a<br />
Figure 18: The operational and functional of Prototype DSP hardware<br />
Figure 18. The operational and functional of Prototype DSP hardware.<br />
(millisecond) and T offset is 4.93 ms (millisecond).<br />
REFERENCES<br />
Badilini F, Erdem T, Zareba W, Moss AJ (2005). ECG Scan: a method<br />
for conversion of paper electrocardiographic printouts to digital<br />
electrocardiographic files, 38: 310-318.<br />
Christov I, Dotsinsky I, Simova I, Prokopova R, Trendafilova E,<br />
Naydenov S (2006). Dataset of manually measured QT intervals in<br />
the electrocardiogram, 5: 31<br />
Christov II S (2006). Fully Automated Method for QT Interval<br />
Measurement in ECG, 33: 321-324.<br />
Gorup Ž, Štajer D, Noč M (2000). Signal Analysis Systems for Optimal<br />
Timing of Electrical Defibrillation, 2: 694-697.<br />
Kaplan AV, Maim DS, Smith JJ, Feigal DA, Simons M, Jefferys D,<br />
Fogarty TJ, Kuntz RE, Leon MB (2004). Medical Device<br />
Development: From Prototype to Regulatory Approval, 109: 3068-<br />
3072.<br />
Kara S, Kemaloğlu S, Kirbaş Ş (2006). Low Cost Compact ECG With<br />
Graphic LCD and Phonocardiogram System Design, 30: 205-209.<br />
Kligfield P, Hancock WE, Helfenbein D E, Dawson JE, Cook AM,<br />
Lindauer MJ, Zhou HS, Xue J (2006). Relation of QT Interval<br />
Measurement to Evolving Automated Algorithm from Different<br />
Manufacturers of Electrocardiographs, 98: 88-92.<br />
b<br />
Analog Front End device<br />
c<br />
e<br />
f<br />
g<br />
Laguna P, Thakor NV, Caminal P, Jane R, Yoon H R (1990). New<br />
algorithm for QT interval analysis in 24 hour Holter ECG: Perform.<br />
Appl., 28: 67-73.<br />
Malarvili MB, Hussain S, Ab-Rahman AR (2005). Development of<br />
Automated 12 Lead QT Dispersion Algorithm for Sudden Cardiac<br />
Death, 2: 2.<br />
Malik M (2004). Errors and Misconceptions in ECG Measurement Used<br />
for the Detectection of Drug Induced QT Interval Prolongation, 37:<br />
25-33.<br />
Mneimneh MA, Povinelli, Johnson MT (2006). Integrative Technique for<br />
the Determination of QT Interval, 33: 329-332.<br />
Mohammad S, Lim Z, Lin Y, Sani A, Zanjani K, Paracha M, Jenkins J<br />
(2003). Real time ECG Analysis Using a TI TMSC54X Digital Signal<br />
Processing Chip, 30: 541-544.<br />
Moraes J, Viana S (1997).A New Automatic Method to Estimate the QT<br />
Interval from the Electrographic Signal, 24: 493-496.<br />
Najeb JM (2007). Real-Time Implementation of Twelve-Lead Automated<br />
Electrocardiogram System Measurement for QT Dispersion Analysis.<br />
Univerisiti Teknologi Malaysia: Masters Thesis.<br />
Najeb JM, Ruhullah A, Salleh Sh H (2005). 12 Channel USB Data<br />
Acquisition System For QT Dispersion Analysis. pp. 83-86.<br />
Online available, Physikalisch-Technische Bundesanstalt (PTB) ECG<br />
data base. URL, http://www.physionet.org/cgi-bin/atm/ATM<br />
Oweis R, Hijazi L (2006). A Computer aided ECG diagnostic tool, 81:<br />
279-284.<br />
Sandham W, Hamilton D, Laguna P, Cohen M (2007). Advances in
Electrocardiogram Signal Processing and Analysis. 2007: 1-5.<br />
Saviola J (2005). The FDA’s role in medical device clinical studies of<br />
human subjects. S1-S4.<br />
Seng et al. 1829<br />
Soon IY, Yeo CK, Ng HC (2001). An analogue video interface for<br />
general purpose DSP. 25: 33-39.
Scientific Research and Essays Vol. 7(18), pp. 1830-1834, 16 May, 2012<br />
Available online at http://www.academicjournals.org/SRE<br />
DOI: 10.5897/SRE11.2187<br />
ISSN 1992-2248 © 2012 <strong>Academic</strong> <strong>Journals</strong><br />
Full Length Research Paper<br />
Effective techniques in drilling to improve the recovery<br />
process of hydrocarbons in oil and gas sector<br />
Shafqat Hameed 1 and Muhammad Abbas Choudhary 2<br />
1 National University of Science and Technology (NUST), Islamabad, Pakistan.<br />
2 University of Engineering and Technology (UET), Taxila, Pakistan<br />
Accepted 25 April, 2012<br />
The purpose of this research is to use effective method of drilling to improve the recovery process of<br />
hydrocarbons in oil and gas sector. Hydrocarbons are present in large amount beneath the earth. By<br />
using exploration geo physics, the reservoir of hydrocarbons below the earth can be located with more<br />
precision and accuracy. The first step consists of the geological and the seismic survey to obtain the<br />
image of the beneath rocks. In the next, step drilling is performed to reach the desired depth.<br />
Depending on the types of rocks beneath the earth the drilling performed may be directional, vertical or<br />
horizontal. This paper will focus on “how multilateral/horizontal drilling can improve the recovery of low<br />
permeable hydrocarbons”. Now a day’s offshore drilling is also performed.<br />
Key words: Hydrocarbons, seismic survey, multilateral drilling, offshore drilling.<br />
INTRODUCTION<br />
Hydrogen and carbon forms hydrocarbon which is an<br />
organic compound in which Crude oil is the most natural<br />
form of hydrocarbon (Clayden and Greeves, 2001).<br />
Hydrocarbon produces lot of energy when burnt and are<br />
the main source of electric energy and home heating.<br />
Hydrocarbons extracted may be either in liquid or<br />
gaseous form such as petroleum or natural gas. To<br />
gather economic values of petroleum, many important<br />
geological fundamentals and processes are needed<br />
(Magoon, 1994). Seismic methods are then used to find<br />
the structure of the layers beneath the earth. Then by<br />
using information from the structural layers, the point of<br />
drilling is defined. First oil well was drilled by Col. Edwin<br />
Drake in USA on August 26, 1859. The well depth was 21<br />
m only and it took most of the summer to reach the<br />
desired depth. (Magoon and Beaumont, 1999)<br />
To recover the hydrocarbons drilling is performed. But<br />
sometimes the usual drilling methods cannot be applied<br />
due to certain hurdles present beneath the Earth and<br />
different location of the reservoirs. The difficulty and cost<br />
*Corresponding author. E-mail:<br />
shafqat.hameed@ceme.nust.edu.pk,<br />
hameed.shafqat@yahoo.com.<br />
associated becomes very high to reach the desire<br />
reservoirs by conventional methods. So further drilling<br />
cannot be performed by using the conventional drilling<br />
methods. Here technology comes to rescue. The<br />
effective techniques employ directional, horizontal, and<br />
multilateral drilling (Figure 1).<br />
Oil drilling is the most common source of recovery of<br />
Petroleum. According to Guerriero (2011), “The step<br />
performed after the structural Geology is reservoir<br />
characterization (mainly in terms of porosity and<br />
permeable structures)”. The main sources of Petroleum<br />
(fossil fuel) are fossilized organic zooplankton and algae<br />
(Kvenvolden, 2006). They were settled at the bottom of<br />
sea and lakes mixing with sediments under anoxic<br />
conditions.<br />
When adequate thermal energy is passed on to the<br />
sedimentary organic matter to break chemical bonds,<br />
petroleum is produced from source rocks. The petroleum<br />
then starts to migrate along the fault zones. Certain<br />
numbers of traps are present in the layer of earth where<br />
petroleum starts to accumulate. They are called reservoir<br />
rocks. Further movement of petroleum is then prevented.<br />
These formations are located with the help of geological<br />
or seismic survey.<br />
When a certain area of interest has been designated, a<br />
survey grid will be drawn up with enough detail to allow
Figure 1. Structure of Methane. Source: (Bowling, 1986).<br />
the interpretation team of geophysicists and geologists to<br />
map the subsurface strata. A survey ship, <strong>complete</strong> with<br />
navigation, recording and shooting equipment (and any<br />
other surveys which may be run simultaneously such as<br />
gravity and magnetic), will then follow this grid recording<br />
the seismic data on computer-compatible tapes for<br />
further processing.<br />
Seismic methods use two phenomenon, reflection and<br />
refraction. They are based on Snell's law. The situation is<br />
directly analogous to optics, where series of shock waves<br />
move in the form of wave fronts from the point of energy<br />
release. When one of these wave fronts encounters an<br />
interface of two dissimilar materials a process of<br />
reflection and refraction takes place. From this survey,<br />
the point of drilling of an oil/gas well is located.<br />
Vertical drilling<br />
Vertical drilling is traditional type of drilling in oil and gas<br />
drilling industry. Drilling a Vertical well is usually cheaper<br />
than drilling a horizontal well. The production obtained<br />
from vertical wells is lesser as compared to the horizontal<br />
wells. The limitation of vertical well is that frequency of<br />
intersecting a large number of fractures is reduced so<br />
less production is obtained.<br />
Horizontal drilling<br />
Horizontal drilling is the same as vertical drilling until the<br />
“kickoff point” which is located just above the target oil or<br />
gas reservoir. From that point, the well is deviated from<br />
the vertical direction to horizontal. Some limitations of<br />
horizontal drilling are:<br />
(a) Horizontal wells have a greater footprint compare to<br />
multilateral wells.<br />
(b) When large number of pay zones are present, more<br />
than one horizontal well are required which is very costly.<br />
Multilateral drilling<br />
Hameed and Choudhary 1831<br />
In this type of drilling, more than one branch emerge from<br />
a single mother wellbore. The branch may be horizontal,<br />
vertical or inclined depending upon the availability of the<br />
zones. More zones can be perforated through multi<br />
lateral drilling from a single mother well.<br />
Directional drilling<br />
It is an angle drilling. Drill pipe provides mechanical and<br />
hydraulic connection between drilling rig at surface and<br />
down hole directional steering system. Drill pipe provides<br />
axial load and disseminate the borehole by destroying the<br />
rocks. Drill pipe is then pumped with the mud fluid cool<br />
and lubricate the rock destruction process and to<br />
transport the rock cutting to the surface (Bowling, 1986).<br />
Plasma channel drilling: According to Martin (1960),<br />
“Plasma channel drilling is a process in which sub micro<br />
second electrical breakdown of rocks is used for efficient<br />
fragmentation of rock Formations. This preferential<br />
electrical breakdown of rock is achieved by the use of a<br />
dielectric liquid (water for example) as the drilling fluid.<br />
The drilling fluid in the PCD process serves as a superior<br />
electrical insulator due to differences in the electrical<br />
properties between the rock and the dielectric liquid<br />
under impulse conditions. The peak power generated at<br />
the drill head is typically hundreds of MW. Such high<br />
powers enable pressures of several GPa to be developed<br />
in the breakdown channel (Figure 2). (Braun and<br />
Burnham, 1993).<br />
METHODOLOGY<br />
The aim of the study is to check how effective techniques in drilling<br />
can improve the recovery of hydrocarbons. The research is carried<br />
out to see whether multilateral drilling can help in the improvement<br />
or not. In order to carry out the survey, 10 to 15 persons working in<br />
oil and gas companies are selected including:
1832 Sci. Res. Essays<br />
1. General Manager drilling operations and services<br />
2. Manager drilling<br />
3. Senior drilling superintendent<br />
4. Geological survey experts.<br />
Figure 2. Drilling method Source: (Bowling, 1986).<br />
A qualitative analysis was carried out to determine the cost<br />
effectiveness of the drilling techniques, the time efficiency and the<br />
limitations associated with them. It also put light on the difficulties<br />
and the specific requirements for the drilling techniques which are<br />
very effective but cannot be employed. The advantages of the latest<br />
technology over the conventional methods are also highlighted.<br />
The analysis consisted of qualitative technique. The search will<br />
include the expert review and cognitive interviews. The result<br />
obtained from this search will be carried out to draw the final<br />
conclusions.<br />
RESULTS<br />
The purpose of this research was that multilateral drilling<br />
can improve the recovery of hydrocarbons. Views were<br />
taken from different persons working in different<br />
organizations. The results are subsequently.<br />
Environmental effect<br />
There are certain wastes associated with the drilling<br />
process. The waste includes the cuttings of the<br />
formations and the chemicals used in mud. In order to<br />
check the effect on environment, 15 persons were<br />
chosen to take their views. Out of which 10 were of the<br />
view of thought that multi lateral drilling has less impact<br />
on the environment, 3 said that it has no effect and 2 said<br />
that the waste is increased (Figure 3).<br />
Time effectiveness<br />
In order to analyze the time associated with the drilling<br />
process, different persons were chosen. Some were of<br />
the opinion that time is reduced because single<br />
foundation is required and dismantling time is also saved.<br />
However, few said that time taken during the drilling is<br />
greater because of the complexity. The results are shown<br />
in Figure 4.<br />
Cost effectiveness<br />
In order to check the cost effectiveness, 5 options were<br />
given to the persons from strongly agree to strongly<br />
disagree. A rating of 5 was given to highest level and 1<br />
was given to the lowest. The results are then drawn from
No of views<br />
the ratings (Figure 5).<br />
Conclusion<br />
No. of views<br />
Figure 3. Effect of waste.<br />
10<br />
5<br />
0<br />
Time effectiveness<br />
Figure 4. Time effectiveness.<br />
Less time greater Greater Time time No effect No effect<br />
From the views of different persons working in different<br />
organizations, the following conclusions can be deducted:<br />
(1) In multilateral wells, higher production is obtained as<br />
compared to conventional wells. Conventional wells<br />
(Vertical Wells) yield smaller contact with the reservoir<br />
when smaller pools are considered, however when<br />
several branches are laid down from the single mother<br />
bore, greater contact with the reservoir is achieved.<br />
Hence high production is obtained.<br />
(2) Multilateral wells are time and cost effective. For<br />
multilateral wells, single foundation for the rig is required.<br />
Conventional wells require separate foundation at every<br />
place. So the time and cost involved in the dismantling,<br />
shifting, joining of the rig and the preparation of the<br />
Hameed and Choudhary 1833<br />
foundation on the next place is saved. Up to the kick off<br />
point a single mother well is drilled, it also saves lot of<br />
cost.<br />
(3) Environmental impacts are decreased in multilateral<br />
wells. Different fluids used in drilling process carry out the<br />
cuttings of drilling. The value of cuttings is reduced in<br />
multilateral wells so as the drilling fluids. Drilling fluids are<br />
made up of different chemicals. So when the cuttings and<br />
fluids are reduced the impact on the environment is also<br />
reduced. This is a big advantage of multilateral drilling.<br />
(4) Quick recovery through multilateral wells. Maximum<br />
number of reservoirs is in contact with multilateral drilling,<br />
so quick recovery of hydrocarbons is obtained as<br />
compared to conventional wells. So payback period is<br />
also smaller.<br />
(5) Less effect on water underneath the earth surface. In<br />
multilateral wells, only a single mother bore well is drilled.<br />
However in conventional drilling method more number of<br />
wells are drilled. Since mud is used for the cleaning of<br />
hole continuously. It is made up of different chemicals so
1834 Sci. Res. Essays<br />
Ratings<br />
Ratings<br />
6<br />
5<br />
4<br />
3<br />
2<br />
1<br />
0<br />
Cost Effectiveness<br />
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15<br />
Figure 5. Cost effectiveness.<br />
the effect of multilateral wells on the water beneath the<br />
earth is reduced.<br />
(6) It also improves the recovery of low permeable gases<br />
which cannot be recovered through conventional<br />
methods.<br />
LIMITATIONS AND RECOMMENDATIONS<br />
There are certain limitations associated with multilateral<br />
drilling. However, certain recommendations will be made<br />
to enhance the usage of multilateral drilling.<br />
(1) Multilateral wells are very complex. They are difficult<br />
to drill. To cope with the complexity, companies must<br />
employ learning of their workers. Proper training should<br />
be given to them by hiring experts in that field.<br />
(2) In multilateral wells, casing is performed only of<br />
mother bore well, however the branches are open.<br />
Controlling of the formations is difficult. So drilling should<br />
be performed only for the developmental wells not for the<br />
exploratory wells.<br />
(3) The equipment used for the multilateral drilling is very<br />
costly. At first glance, this technology looks to be very<br />
expensive. However, the overall advantage should be<br />
considered which is much greater when production is<br />
obtained.<br />
(4) Well control becomes very difficult when production<br />
from different branches is obtained. However, by using<br />
Christmas tree (equipment having separate well heads),<br />
the problem can be overcome. The production coming<br />
from branches will go to separate heads.<br />
Persons<br />
In summary, the conventional drilling is not much<br />
complex to perform and that is why it is popular in<br />
Pakistan. But the production obtained is not in greater<br />
quantity. On the other hand, multilateral wells are difficult<br />
to drill and the equipment is expensive but the advantage<br />
becomes greater when production is obtained. So this<br />
should be preferred. In Pakistan, no company has<br />
performed multilateral drilling yet but work is going on for<br />
the adaptation of this technology. It has bright future in<br />
Pakistan provided that government of Pakistan<br />
emphasizes its use and invests heavily in this technology.<br />
REFERENCES<br />
Bowling JP, Jogi PN, Burgess TM (1986). “Three Dimensional Bottom<br />
hole Assembly Model Improves Directional Drilling,” IADC/SPE Conf.,<br />
SPE 14768, Dallas, Feb.<br />
Braun, Robert L, Burnham, Lan K (1993). “Chemical Reaction Model for<br />
Oil and Gas Generation from Type I and Type II Kerogen.<br />
Clayden J, Greeves N (2001) Organic Chemistry Oxford ISBN<br />
0198503466, p. 21.<br />
Guerriero V (2011). “Improved statistical multi-scale analysis of<br />
fractures in carbonate reservoir analogues”<br />
Kvenvolden KA (2006). “Organic geochemistry – A retrospective of its<br />
first 70 years”. Organ. Geochem., p. 37.<br />
Magoon LB, Beaumont EA (1999). Petroleum Systems. In: Handbook of<br />
Petroleum Geology: Exploring for Oil and Gas Traps, Beaumont,<br />
E.A.; Foster, H.N. (Eds), American Association of Petroleum<br />
Geologists Tulsa, pp. 3-1, 3-34.<br />
Magoon LB, Dow, WG (Eds.) (1994). The Petroleum System-From<br />
Source to Trap, AAPG Memoir. American Association of Petroleum<br />
Geologists, Tulsa, 60: 655.<br />
Martin E (1960). “Experimental investigation of a high energy, high<br />
pressure or plasma”, J. Appl. Phys., 31: 255-267.
UPCOMING CONFERENCES<br />
ICPP 2012: International Conference on Plasma Physics Venice, Italy<br />
November 14-16, 2012<br />
PRESCO 2012 will be held at the Crowne Plaza Hotel<br />
Hiroshima, Japan August 6th – 9th, 2012
Conferences and Advert<br />
July 2012<br />
5th International Conference on Integrated Systems, Design and Technology (ISDT),<br />
Kuala Lumpur, Malaysia, 6 Jul 2012<br />
36th Annual Scientific Meeting of The Human Genetics Society of Australasia<br />
(HGSA), Canberra, Australia, 22 Jul 2012<br />
Yeast Genetics Conference, Princeton, USA, 31 Jul 2012<br />
August 2012<br />
The Inaugural Pacific Rim Energy & Sustainability Congress, Hiroshima, Japan, 6<br />
Aug 2012<br />
November 2012<br />
4th International Conference on Science in Society, Berkeley, USA, 15 Nov 2012
Scientific Research<br />
and Essays<br />
Related <strong>Journals</strong> Published by <strong>Academic</strong> <strong>Journals</strong><br />
■ Journal of Oceanography and Marine Science<br />
■ Journal of Internet and Information Systems<br />
■ African Journal of Pure and Applied Chemistry<br />
■ International Journal of the Physical Sciences<br />
■ Journal of Geology and Mining Research