Download complete issue (pdf 3360kb) - Academic Journals

Scientific Research 

and Essays 

Volume 7 Number 18 16 May, 2012 

ISSN 1992-2248

ABOUT SRE 

The Scientific Research and Essays (SRE) is published weekly (one volume per year) by Academic Journals. 

Scientific Research and Essays (SRE) is an open access journal with the objective of publishing quality research 

articles in science, medicine, agriculture and engineering such as Nanotechnology, Climate Change and Global 

Warming, Air Pollution Management and Electronics etc. All papers published by SRE are blind peer reviewed. 

Submission of Manuscript 

Submit manuscripts as e-mail attachment to the Editorial Office at: sre@cademicjournals.org, 

sre.research.journal@gmail.com. A manuscript number will be mailed to the corresponding author shortly 

after submission. 

For all other correspondence that cannot be sent by e-mail, please contact the editorial office for the 

appropriate address. 

The Scientific Research and Essays will only accept manuscripts submitted as e-mail attachments. 

Please read the Instructions for Authors before submitting your manuscript. The manuscript files should be 

given the last name of the first author.

Editors 

Dr. NJ Tonukari 

Editor-in-Chief 

Scientific Research and Essays 

Academic Journals 

E-mail: sre.research.journal@gmail.com 

Dr. M. Sivakumar Ph.D. (Tech). 

Associate Professor 

School of Chemical & Environmental Engineering 

Faculty of Engineering 

University of Nottingham 

Jalan Broga, 43500 Semenyih 

Selangor Darul Ehsan 

Malaysia. 

Prof. N. Mohamed El Sawi Mahmoud 

Department of Biochemistry, Faculty of science, 

King AbdulAziz university, 

Saudia Arabia. 

Prof. Ali Delice 

Science and Mathematics Education Department, 

Atatürk Faculty of Education, 

Marmara University, 

Turkey. 

Prof. Mira Grdisa 

Rudjer Boskovic Institute, Bijenicka cesta 54, 

Croatia. 

Prof. Emmanuel Hala Kwon-Ndung 

Nasarawa State University Keffi Nigeria 

PMB 1022 Keffi, 

Nasarawa State. 

Nigeria. 

Dr. Cyrus Azimi 

Department of Genetics, Cancer Research Center, 

Cancer Institute, Tehran University of Medical Sciences, 

Keshavarz Blvd., 

Tehran, Iran. 

Dr. Gomez, Nidia Noemi 

National University of San Luis, 

Faculty of Chemistry, Biochemistry and Pharmacy, 

Laboratory of Molecular Biochemistry Ejercito de los 

Andes 950 - 5700 San Luis 

Argentina. 

Prof. M. Nageeb Rashed 

Chemistry Department- Faculty of Science, Aswan 

South Valley University, 

Egypt. 

Dr. John W. Gichuki 

Kenya Marine & Fisheries Research Institute, 

Kenya. 

Dr. Wong Leong Sing 

Department of Civil Engineering, 

College of Engineering, 

Universiti Tenaga Nasional, 

Km 7, Jalan Kajang-Puchong, 

43009 Kajang, Selangor Darul Ehsan, 

Malaysia. 

Prof. Xianyi LI 

College of Mathematics and Computational Science 

Shenzhen University 

Guangdong, 518060 

P.R. China. 

Prof. Mevlut Dogan 

Kocatepe University, Science Faculty, 

Physics Dept. Afyon/ Turkey. 

Turkey . 

Prof. Kwai-Lin Thong 

Microbiology Division, 

Institute of Biological Science, 

Faculty of Science,University of Malaya, 

50603, Kuala Lumpur, 

Malaysia. 

Prof. Xiaocong He 

Faculty of Mechanical and Electrical Engineering, 

Kunming University of Science and Technology, 

253 Xue Fu Road, Kunming, 

P.R. China. 

Prof. Sanjay Misra 

Department of Computer Engineering 

School of Information and Communication Technology 

Federal University of Technology, Minna, 

Nigeria. 

Prof. Burtram C. Fielding Pr.Sci.Nat. 

Department of Medical BioSciences 

University of the Western Cape 

Private Bag X17 

Modderdam Road 

Bellville, 7535, 

South Africa. 

Prof. Naqib Ullah Khan 

Department of Plant Breeding and Genetics 

NWFP Agricultural University Peshawar 25130, 

Pakistan

Editorial Board 

Prof. Ahmed M. Soliman 

20 Mansour Mohamed St.,Apt 51, 

Zamalek, Cairo, 

Egypt. 

Prof. Juan José Kasper Zubillaga 

Av. Universidad 1953 Ed. 13 depto 304, 

México D.F. 04340, 

México. 

Prof. Chau Kwok-wing 

University of Queensland 

Instituto Mexicano del Petroleo, 

Eje Central Lazaro Cardenas 

Mexico D.F., 

Mexico. 

Prof. Raj Senani 

Netaji Subhas Institute of Technology, 

Azad Hind Fauj Marg, 

Sector 3, 

Dwarka, New Delhi 110075, 

India. 

Prof. Robin J Law 

Cefas Burnham Laboratory, 

Remembrance AvenueBurnham on Crouch, 

Essex CM0 8HA, 

UK. 

Prof. V. Sundarapandian 

Indian Institute of Information Technology and 

Management-Kerala 

Park Centre, 

Technopark Campus, Kariavattom P.O., 

Thiruvananthapuram-695 581, Kerala, 

India. 

Prof. Tzung-Pei Hong 

Department of Electrical Engineering, 

and at the Department of Computer Science and 

Information Engineering 

National University of Kaohsiung. 

Prof. Zulfiqar Ahmed 

Department of Earth Sciences, box 5070, 

Kfupm,dhahran - 31261, 

Saudi Arabia. 

Prof. Khalifa Saif Al-Jabri 

Department of Civil and Architectural Engineering 

College of Engineering, 

Sultan Qaboos University 

P.O. Box 33, Al-Khod 123, Muscat. 

Sultanate of Oman. 

Prof. Konstantinos 

43 Liakateon St.11474 Athens, 

Greece 

Prof. Liotta Leonarda 

Prof. V. Sundarapandian 

Indian Institute of Information Technology & Management - 

Kerala 

Park Centre, 

Technopark, Kariavattom P.O. 

Thiruvananthapuram-695 581, 

Kerala India. 

Prof. Thangavelu Perianan 

Department of Mathematics, Aditanar College, 

Tiruchendur-628216 India. 

Prof. Yan-ze Peng 

Department of Mathematics, 

Huazhong University of Science and Technology, 

Wuhan 430074, P. R. 

China. 

Prof. Konstantinos D. Karamanos 

Universite Libre de Bruxelles, 

CP 231 Centre of Nonlinear Phenomena 

And Complex systems, 

CENOLI Boulevard de Triomphe 

B-1050, 

Brussels, Belgium. 

Prof. Xianyi LI 

School of Mathematics and Physics, 

Nanhua University, Hengyang City, 

Hunan Province, 

P. R. China. 

Dr. K.W. Chau 

Hong Kong Polytechnic University 

Department of Civil & Structural Engineering, 

Hong Kong Polytechnic University, Hunghom, 

Kowloon, Hong Kong, 

China. 

Dr. Amadou Gaye 

LPAO-SF / ESP Po Box 5085 Dakar-Fann SENEGAL 

University Cheikh Anta Diop Dakar 

SENEGAL. 

Prof. Masno Ginting 

P2F-LIPI, Puspiptek-Serpong, 

15310 Indonesian Institute of Sciences, 

Banten-Indonesia. 

Dr. Ezekiel Olukayode Idowu 

Department of Agricultural Economics, 

Obafemi Awolowo University, 

Ife-Ife, 

Nigeria. 

Dr. Fei Kang 

Faculty of Infrastructure Engineering, 

Dalian University of Technology, 

Dalian, 116024, China 

Dr. Jane Wanjiku Kabubo-Mariara

Electronic submission of manuscripts is strongly 

encouraged, provided that the text, tables, and figures are 

included in a single Microsoft Word file (preferably in Arial 

font). 

The cover letter should include the corresponding author's 

full address and telephone/fax numbers and should be in 

an e-mail message sent to the Editor, with the file, whose 

name should begin with the first author's surname, as an 

attachment. 

Article Types 

Three types of manuscripts may be submitted: 

Regular articles: These should describe new and carefully 

confirmed findings, and experimental procedures should 

be given in sufficient detail for others to verify the work. 

The length of a full paper should be the minimum required 

to describe and interpret the work clearly. 

Short Communications: A Short Communication is suitable 

for recording the results of complete small investigations 

or giving details of new models or hypotheses, innovative 

methods, techniques or apparatus. The style of main 

sections need not conform to that of full-length papers. 

Short communications are 2 to 4 printed pages (about 6 to 

12 manuscript pages) in length. 

Reviews: Submissions of reviews and perspectives covering 

topics of current interest are welcome and encouraged. 

Reviews should be concise and no longer than 4-6 printed 

pages (about 12 to 18 manuscript pages). Reviews are also 

peer-reviewed. 

Review Process 

Instructions for Author 

All manuscripts are reviewed by an editor and members of 

the Editorial Board or qualified outside reviewers. Authors 

cannot nominate reviewers. Only reviewers randomly 

selected from our database with specialization in the 

subject area will be contacted to evaluate the manuscripts. 

The process will be blind review. 

Decisions will be made as rapidly as possible, and the 

journal strives to return reviewers’ comments to authors as 

fast as possible. The editorial board will re-review 

manuscripts that are accepted pending revision. It is the 

goal of the SRE to publish manuscripts within weeks after 

submission. 

Regular articles 

All portions of the manuscript must be typed doublespaced 

and all pages numbered starting from the title 

page. 

The Title should be a brief phrase describing the 

contents of the paper. The Title Page should include the 

authors' full names and affiliations, the name of the 

corresponding author along with phone, fax and E-mail 

information. Present addresses of authors should 

appear as a footnote. 

The Abstract should be informative and completely selfexplanatory, 

briefly present the topic, state the scope of 

the experiments, indicate significant data, and point out 

major findings and conclusions. The Abstract should be 

100 to 200 words in length.. Complete sentences, active 

verbs, and the third person should be used, and the 

abstract should be written in the past tense. Standard 

nomenclature should be used and abbreviations should 

be avoided. No literature should be cited. 

Following the abstract, about 3 to 10 key words that will 

provide indexing references should be listed. 

A list of non-standard Abbreviations should be added. 

In general, non-standard abbreviations should be used 

only when the full term is very long and used often. 

Each abbreviation should be spelled out and introduced 

in parentheses the first time it is used in the text. Only 

recommended SI units should be used. Authors should 

use the solidus presentation (mg/ml). Standard 

abbreviations (such as ATP and DNA) need not be 

defined. 

The Introduction should provide a clear statement of 

the problem, the relevant literature on the subject, and 

the proposed approach or solution. It should be 

understandable to colleagues from a broad range of 

scientific disciplines. 

Materials and methods should be complete enough 

to allow experiments to be reproduced. However, only 

truly new procedures should be described in detail; 

previously published procedures should be cited, and 

important modifications of published procedures should 

be mentioned briefly. Capitalize trade names and 

include the manufacturer's name and address. 

Subheadings should be used. Methods in general use 

need not be described in detail.

Results should be presented with clarity and precision. 

The results should be written in the past tense when 

describing findings in the authors' experiments. 

Previously published findings should be written in the 

present tense. Results should be explained, but largely 

without referring to the literature. Discussion, 

speculation and detailed interpretation of data should 

not be included in the Results but should be put into the 

Discussion section. 

The Discussion should interpret the findings in view of 

the results obtained in this and in past studies on this 

topic. State the conclusions in a few sentences at the end 

of the paper. The Results and Discussion sections can 

include subheadings, and when appropriate, both 

sections can be combined. 

The Acknowledgments of people, grants, funds, etc 

should be brief. 

Tables should be kept to a minimum and be designed to 

be as simple as possible. Tables are to be typed doublespaced 

throughout, including headings and footnotes. 

Each table should be on a separate page, numbered 

consecutively in Arabic numerals and supplied with a 

heading and a legend. Tables should be self-explanatory 

without reference to the text. The details of the methods 

used in the experiments should preferably be described 

in the legend instead of in the text. The same data should 

not be presented in both table and graph form or 

repeated in the text. 

Figure legends should be typed in numerical order on a 

separate sheet. Graphics should be prepared using 

applications capable of generating high resolution GIF, 

TIFF, JPEG or Powerpoint before pasting in the Microsoft 

Word manuscript file. Tables should be prepared in 

Microsoft Word. Use Arabic numerals to designate 

figures and upper case letters for their parts (Figure 1). 

Begin each legend with a title and include sufficient 

description so that the figure is understandable without 

reading the text of the manuscript. Information given in 

legends should not be repeated in the text. 

References: In the text, a reference identified by means 

of an author‘s name should be followed by the date of 

the reference in parentheses. When there are more than 

two authors, only the first author‘s name should be 

mentioned, followed by ’et al‘. In the event that an 

author cited has had two or more works published during 

the same year, the reference, both in the text and in the 

reference list, should be identified by a lower case letter 

like ’a‘ and ’b‘ after the date to distinguish the works. 

Examples: 

Abayomi (2000), Agindotan et al. (2003), (Kelebeni, 

1983), (Usman and Smith, 1992), (Chege, 1998; 

Chukwura, 1987a,b; Tijani, 1993, 1995), (Kumasi et al., 

2001) 

References should be listed at the end of the paper in 

alphabetical order. Articles in preparation or articles 

submitted for publication, unpublished observations, 

personal communications, etc. should not be included 

in the reference list but should only be mentioned in 

the article text (e.g., A. Kingori, University of Nairobi, 

Kenya, personal communication). Journal names are 

abbreviated according to Chemical Abstracts. Authors 

are fully responsible for the accuracy of the references. 

Examples: 

Ogunseitan OA (1998). Protein method for 

investigating mercuric reductase gene expression in 

aquatic environments. Appl. Environ. Microbiol. 

64:695–702. 

Gueye M, Ndoye I, Dianda M, Danso SKA, Dreyfus B 

(1997). Active N2 fixation in several Faidherbia albida 

provenances. Ar. Soil Res. Rehabil. 11:63-70. 

Charnley AK (1992). Mechanisms of fungal 

pathogenesis in insects with particular reference to 

locusts. In: Lomer CJ, Prior C (eds) Biological Controls of 

Locusts and Grasshoppers: Proceedings of an 

international workshop held at Cotonou, Benin. Oxford: 

CAB International, pp 181-190. 

Mundree SG, Farrant JM (2000). Some physiological 

and molecular insights into the mechanisms of 

desiccation tolerance in the resurrection plant 

Xerophyta viscasa Baker. In Cherry et al. (eds) Plant 

tolerance to abiotic stresses in Agriculture: Role of 

Genetic Engineering, Kluwer Academic Publishers, 

Netherlands, pp 201-222. 

Babalola OO (2002). Interactions between Striga 

hermonthica (Del.) Benth. and fluorescent rhizosphere 

bacteria Of Zea mays, L. and Sorghum bicolor L. 

Moench for Striga suicidal germination In Vigna 

unguiculata . PhD dissertation, University of Ibadan, 

Ibadan, Nigeria. 

Short Communications 

A Short Communication is suitable for recording the 

results of complete small investigations or giving 

details of new models or hypotheses, innovative 

methods, techniques or apparatus. The style of main 

sections need not conform to that of full-length papers. 

Short communications are 2 to 4 printed pages (about 

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 

and Essays is not contingent upon the author's ability to pay the charges. Neither is acceptance to pay the handling 

fee a guarantee that the paper will be accepted for publication. Authors may still request (in advance) that the 

editorial office waive some of the handling fee under special circumstances. 

Copyright: © 2012, Academic Journals. 

All rights Reserved. In accessing this journal, you agree that you will access the contents for your own personal use 

but not for any commercial use. Any use and or copies of this Journal in whole or in part must include the customary 

bibliographic citation, including author attribution, date and article title. 

Submission of a manuscript implies: that the work described has not been published before (except in the form of an 

abstract or as part of a published lecture, or thesis) that it is not under consideration for publication elsewhere; that if 

and when the manuscript is accepted for publication, the authors agree to automatic transfer of the copyright to the 

publisher. 

Disclaimer of Warranties 

In no event shall Academic Journals be liable for any special, incidental, indirect, or consequential damages of any 

kind arising out of or in connection with the use of the articles or other material derived from the SRE, whether or not 

advised of the possibility of damage, and on any theory of liability. 

This publication is provided "as is" without warranty of any kind, either expressed or implied, including, but not 

limited to, the implied warranties of merchantability, fitness for a particular purpose, or non-infringement. 

Descriptions of, or references to, products or publications does not imply endorsement of that product or publication. 

While every effort is made by Academic Journals to see that no inaccurate or misleading data, opinion or statements 

appear in this publication, they wish to make it clear that the data and opinions appearing in the articles and 

advertisements herein are the responsibility of the contributor or advertiser concerned. Academic Journals makes no 

warranty of any kind, either express or implied, regarding the quality, accuracy, availability, or validity of the data or 

information in this publication or of any other publication to which it may be linked.

International Journal 

Scientific 

of Medicine 

Research 

and 

and 

Medical 

Essays 

Sciences 

Table of Contents: Volume 7 Number 18 16 May, 2012 

es 

ARTICLES 

Gastroprotective effects of Dicranopteris linearis leaf extract against 

ethanol-induced gastric mucosal injury in rats 1761 

Jamal Hussaini, Nurul Asyikin Othman, Mahmood Ameen Abdulla, Nazia 

Abdul Majid, Halabi Mohd Faroq and Salmah Ismail 

Introducing slag powder as drag reduction agent in pipeline: An 

experimental approach 1768 

Hayder A. Abdulbari, Siti Nuraffini Bt Kamarulizam, Rosli M. Y. and 

Arun Gupta 

Available transfer capability and least square method 1777 

Mojgan Hojabri, Mohammadsoroush Soheilirad and Mahdi Hedayati 

Seasonality of non-timber forest products in the Kupe mountain region 

of South Cameroon 1786 

Ngane B. K., Ngane E. B., Sumbele S. A., Njukeng J. N., Ngone M. A. and 

Ehabe E. E. 

Adaptive message size routing strategy for delay tolerant network 1798 

Qaisar Ayub, M. Soperi Mohd Zahid, Sulma Rashid and Abdul Hanan Abdullah 

Length-weight relationships, relative condition factor and relative weight 

of three fish species from beach seine fishing grounds in Iranian coastal 

waters of Caspian Sea 1809 

Moradinasab, GH., Raeisi, H, Paighambari, S. Y., Ghorbani, R and Bibak, Z. 

Design and development of standalone DSP prototype for QT interval 

processing and monitoring 1813 

Goh Chun Seng, Sh-Hussain Salleh, J. M. Najeb, I. Kamarulafizam, 

Mahyar Hamedi and Alias Md Noor 

Effective techniques in drilling to improve the recovery process of 

hydrocarbons in oil and gas sector 1830 

Shafqat Hameed and Muhammad Abbas Choudharyand Alias Md Noor

Scientific Research and Essays Vol. 7(18), pp. 1761-1767, 16 May, 2012 

Available online at http://www.academicjournals.org/SRE 

DOI: 10.5897/SRE11.775 

ISSN 1992-2248 ©2012 Academic Journals 

Full Length Research Paper 

Gastroprotective effects of Dicranopteris linearis leaf 

extract against ethanol-induced gastric mucosal 

injury in rats 

Jamal Hussaini 1 *, Nurul Asyikin Othman 1 , Mahmood Ameen Abdulla 2 , Nazia Abdul Majid 3 , 

Halabi Mohd Faroq 3 and Salmah Ismail 2 

1 Faculty of Medicine, Universiti Teknologi MARA (UiTM), 40450 Shah Alam, Selangor, Malaysia. 

2 Department of Molecular Medicine, Faculty of Medicine, University of Malaya, 50603, Malaysia. 

3 Institute of Biological Science, Faculty of Science, University of Malaya, 50603, Malaysia. 

Accepted 9 May, 2012 

Dicranopteris linearis is a medicinal plant commonly used traditionally in the treatment of many 

ailments. This study was performed to evaluate the gastroprotective effect of ethanolic extracts of D. 

linearis leaf extract (DLELE) against ethanol-induced gastric mucosal injury in experimental rats. The 

rats were divided into four groups respectively pre-treated orally with carboxymethyl cellulose (CMC) 

solution (ulcer control groups), omeprazole 20 mg/kg (reference group), 250 and 500 mg/kg of DLELE 

(experimental groups) one hour before oral administration of absolute ethanol to generate gastric 

mucosal damage. After an additional hour, the rats were sacrificed and the ulcer areas of the gastric 

walls were determined. The ulcer control group exhibited severe mucosal injury, whereas groups pretreated 

with DLELE exhibited significant protection of gastric mucosa. These findings were also 

confirmed by histology of gastric wall. Significant increases in gastric mucus production and decrease 

in acidity of gastric content were observed in treated groups with DLELE compare to ulcer control 

group. In conclusion, treatment with DLELE prior to absolute alcohol has significantly protect gastric 

mucosa as ascertained grossly by significant reduction of ulcer area, increases in gastric mucus 

production and decrease the acidity of gastric content and histology by comparatively decreases in 

gastric mucosal injury, reduction or absence of edema and leucocytes infiltration of submucosal layer 

compared to ulcer control group. DLELE was able to decrease the acidity and increase the mucosal 

defense in the gastric area, thereby justifying its use as an antiulcerogenic agent. 

Key words: Dicranopteris linearis, cytoprotection, gastric ulcer, mucus, histology. 

INTRODUCTION 

Gastric ulcer is an illness that affects a considerable 

number of people worldwide. The etiological factors of 

this disorder include stress, smoking, nutritional 

deficiencies, infections, frequent and indiscriminate use 

of nonsteroidal anti-inflammatory drugs (NSAIDs) 

(Khazaei and Salehi, 2006). The pathogenesis of 

*Corresponding author. E-mail: jamalum2004@yahoo.com, 

jamalh@salam.uitm.edu.my. Tel: +603-55442886. Fax: +603- 

55442831. 

gastroduodenal ulcers are influenced by various 

aggressive and defensive factors, such as mucus 

secretion, mucosal barrier, acid-pepsin secretion, blood 

flow, cellular regeneration and endogenous protective 

agents (prostaglandins and epidermal growth factor 

(Mizui et al., 1987). Although, the introduction of protonpump 

inhibitors to the classic anti-ulcer therapy had 

revolutionized treatment of peptic ulcers and other 

gastrointestinal disorders, there is still no complete cure 

for this disease. It has been shown that long term use of 

these drugs leads to various adverse and side effects. 

Relapses of the malady, ineffectiveness of different drug

1762 Sci. Res. Essays 

regimens and even resistance to drugs are emerging (Al- 

Mofleh et al., 2007). Thus, there is an urgent requirement 

to identify more effective and safe anti-ulcer agents. A 

widespread search has been launched to identify new 

anti-ulcer therapies from natural sources. Herbs, 

medicinal plants, spices, vegetables and crude drug 

substances are considered to be a potential source to 

combat various diseases including gastric ulcer. In the 

scientific literature, a large number of medicinal plants 

with gastric anti-ulcer potential have been reported 

(Abdulla et al., 2010; Ketuly et al., 2011; Mahmood et al., 

2010; Wasman et al., 2010). 

Many plants are being used in the traditional medicine 

because they produce a diverse range of bioactive 

molecules, making them a rich source of different types 

of medicines (Tanaka et al., 2006). Dicranopteris linearis 

(Gleicheniaceae), known locally to the Malay’s as Resam 

has been used in the Malay’s traditional medicine as a 

cooling drink and also to reduce fever (Zakaria et al., 

2006). In other part of the world, it is used to treat asthma 

and for women’s sterility (Vasuda, 1999), and to get rid of 

intestinal worms infection (Chin, 1992). Scientifically, D. 

linearis extracts have been reported to possess antinociceptive, 

anti-inflammatory and antipyretic activities 

(Zakaria et al., 2008), antibacterial activity (Lai et al., 

2009) and potential cytotoxic and antioxidant activity 

against various types of cancer (Zakaria et al., 2011). 

Phytochemical study has revealed the presence of 

various types of flavonoids, particuarly of flavonol 3-Oglycosides 

types, and triterpenes, saponins and high 

content of steroids in the leaves of D. Linearis (Raja et 

al., 1995; Zakaria, 2007). Thus far, there is no data 

available on gastroprotective activity of DLELE. The 

present study was undertaken to evaluate antiulcerogenic 

properties of DLELE in rats. 

MATERIALS AND METHODS 

In this study, omeprazole was used as the reference anti-ulcer 

drug, and was obtained from the University Malaya Medical Centre 

(UMMC) Pharmacy. The drug was dissolved in carboxymethyl 

cellulose (0.5% w/v) (CMC) and administered orally to the rats in 

concentrations of 20 mg/kg body weight (5 ml/kg) according to the 

recommendation of (Abdulla et al., 2010). 

Plant specimen and extract preparation 

D. linearis leaves were obtained from Ethno Resources Sdn Bhd, 

Selangor Malaysia, and identified by comparison with the Voucher 

specimen deposited at the Herbarium of Rimba Ilmu, Institute of 

Science Biology, University of Malaya, Kuala Lumpur. The dried 

leaves were powdered using electrical blender. Hundred grams of 

the fine powder were soaked in 500 ml of 95% ethanol in conical 

flask for 3 days. After 3 days the mixture was filtered using a fine 

muslin cloth followed by filter paper (Whatman No. 1) and distilled 

under reduced pressure in an Eyela rotary evaporator (Sigma- 

Aldrich, USA). The dry extract was then dissolved in CMC(0.5% 

w/v) and administered orally to rats in concentrations of 250 

and 500 mg/kg body weight (5 ml/kg body weight) according to the 

recommendation of (Mahmood et al., 2010). 

Experimental animals for gastric ulcer 

Sprague Dawley healthy adult male rats were obtained from the 

Experimental Animal House, Faculty of Medicine, University of 

Malaya, and Ethic No. PM/27/07/2010/MAA (R). The rats were 

divided randomly into 4 groups of 6 rats each. Each rat that 

weighed between 200 - 225 g was placed individually in a separate 

cage (one rat per cage) with wide-mesh wire bottoms to prevent 

coprophagia during the experiment. The animals were maintained 

on standard pellet diet and tap water. The study was approved by 

the Ethics Committee for Animal Experimentation, Faculty of 

Medicine, University of Malaya, Malaysia. Throughout the 

experiments, all animals received human care according to the 

criteria outlined in the “Guide for the Care and Use of laboratory 

Animals” prepared by the National Academy of Sciences and 

published by the National Institute of Health. 

Gastric ulcer-induction by absolute ethanol 

The rats fasted for 48 h before the experiment (Abdulla et al., 

2010), but were allowed free access to drinking water up till 2 h 

before the experiment. Gastric ulcer was induced by orogastric 

intubation of absolute ethanol (5 ml/kg) according to the method 

described by Mahmood et al. (2010). Ulcer control groups were 

orally administered vehicle (CMC, 0.5% w/v, 5 ml/kg). The 

reference group received oral doses of 20 mg/kg omeprazole in 

CMC (5 ml/kg) as positive control. Experimental groups were orally 

administered DLELE in CMC solution (5 ml/kg) at doses of 250 and 

500 mg/kg . One hour after this pre-treatment all groups of rats 

were administered with absolute ethanol (5 ml/kg) in order to induce 

gastric ulcers (Abdulla et al., 2010). The rats were euthanized 60 

min later (Ketuly et al., 2011) under an overdose of xylazin and 

ketamine anesthesia and their stomachs were immediately excised. 

Measurement of mucus production 

Gastric mucus production was measured in the rats that were 

subjected to absolute ethanol-induced gastric lesions. The gastric 

mucosa of each rat was obtained by gentle scraping the mucosa 

with a glass slide and the collected mucus were weighed by using a 

precision electronic balance (Ketuly et al., 2011; Wasman et al., 

2010). 

Measurement of acid content of gastric juice (pH) 

Samples of gastric contents were analyzed for hydrogen ion 

concentration by pH metric titration with 0.1 N NaOH solutions 

using digital pH meter (Abdulla et al., 2010; Ketuly et al., 2011). 

Gross gastric lesions evaluation 

Ulcers of the gastric mucosa appear as elongated bands of 

hemorrhagic lesions parallel to the long axis of the stomach. Gastric 

mucosa of each rat was thus examined for damage. The length and 

width of the ulcer (mm) were measured by a planimeter (10 × 10 

mm 2 = ulcer area) under dissecting microscope (×1.8). The 

ulcerated area was measured by counting the number of small 

squares, 2 mm × 2 mm, covering the length and width of each ulcer 

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. 

Animal 

group 

Pre-treatment 

(5 ml/kg dose) 

Mucus 

production 

pH of gastric 

content 

Ulcer area (mm) 2 

(Mean ± S.E.M) 

Hussaini et al. 1763 

Inhibition 

(%) 

1 CMC (Ulcer control) 0.37 ± 0.01 a 3.88 ± 0.01 a 955.25 ± 2.82 a - 

2 Omeprazole (20 mg/kg) 0.57 ± 0.01 b 7.14 ± 0.33 b 148.3 � 2.62 b 84.48 

3 DLELE (250 mg/kg) 0.52 ± 0.01 c 5.6 ± 0.1 c 225.11 ± 3.65 c 76.43 

4 DLELE (500 mg/kg) 0.77 ± 0.01 d 6.22 ± 0.01 c 38.08 � 2.18 d 96.01 

All values are expressed as mean ± standard error mean. Means with different superscripts are significantly different. The mean 

difference is significant at the p>0.05 level. 

applied in the calculation of the ulcer area (UA) wherein the sum of 

small squares × 4 × 1.8 = UA (mm 2 ) according to the 

recommendation of Mahmood et al. (2010). The inhibition 

percentage (I.0%) was calculated using the following formula 

according to the recommendation of Wasman et al. (2010). 

(I%) = [(UAcontrol − UAtreated) � UAcontrol] × 100%. 

Histological evaluation of gastric lesions 

Specimens of the gastric walls of each rat were fixed in 10% 

buffered formalin and processed in a paraffin tissue processing 

machine. Sections of the stomach were made at a thickness of 5 

µm and stained with hematoxylin and eosin for histological 

evaluation (Abdulla et al., 2010; Ketuly et al., 2011). 

Statistical analysis 

All values were reported as mean ± S.E.M. Statistical significance 

of differences between groups was assessed using one-way 

ANOVA. A value of p


Figure 1. Gross appearance of the gastric mucosa in rats. (a) pre-treated with 5 ml/kg CMC (ulcer 

control). Severe injuries are seen in the gastric mucosa (arrow). Absolute ethanol produced extensive 

visible hemorrhagic necrosis of gastric mucosa. (b) pre-treated with of omeprazole (20 mg/kg). Injuries to 

the gastric mucosa are very milder compared to the injuries seen in the ulcer control rats (arrow). (c) pretreated 

with DLELE (250 mg/kg). Mild injuries are seen in the gastric mucosa. The extract reduces the 

formation of gastric lesions induced by absolute ethanol (arrow). (d) pre-treated with DLELE 500 mg/kg. 

Mild injuries to the gastric mucosa are seen, and flattening of the gastric mucosa is seen (arrow). 

mucus production. This suggests that gastro-protective 

effect of DLELE is mediated partly by preservation of 

gastric mucus production. 

Oxidative stress plays an important role in the pathogenesis 

of various diseases including gastric ulcer, with 

antioxidants being reported to play a significant role in the 

protection of gastric mucosa against various necrotic 

agents (Trivedi and Rawal, 2001). Administration of 

antioxidants inhibits ethanol-induced gastric injury in rat 

(Ligumsky et al., 1995). DLELE possesses a broad 

spectrum of biological activities, and the plant extract has 

been shown to contain pharmaceutically active chemical 

constituents, such as flavonoids, saponins and 

terpenoids (Zakaria, 2007) and it is speculated that the 

gastroprotective effect exerted by DLELE could be 

attributed to its antioxidant property. Antioxidant property 

of the DLELE may possibly counteract oxidative damage 

caused by absolute ethanol toxicity. The observed 

anti-ulcerogenic activity may be due to its antioxidant 

effects and appears to strengthen the mucosal barrier, 

which is the first line of defense against endogenous and 

exogenous ulcerogenic agents. Previous studies have 

shown that flavonoids may be related to the antiulcer 

activity (Hiruma-Lima et al., 2006), and play a major role 

in the mechanism of gastroprotection (La Casa et al., 

2000). It could be conceivable that the anti-ulcer activity 

of this plant could be linked to the flavonoids since 

flavonoids are reported to protect the mucosa by 

preventing the formation of lesions by various necrotic 

agents (Saurez et al., 1996). It is well known that many 

flavonoids display anti-secretory and cytoprotective 

properties in different experimental models of gastric 

ulcer (Zayachkivska et al., 2005). Flavonoids possess 

anti-oxidant properties in addition to strengthening the 

mucosal defense system through stimulation of gastric 

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 

with 5 ml/kg of CMC (ulcer control). There is severe disruption to the surface epithelium, 

necrotic lesions penetrate deeply into mucosa (orang arrow) and extensive edema of submucosa 

layer and leucocyte infiltration are present (White arrow). (b) pre-treated with omeprazole (20 

mg/kg). Mild disruption of the surface epithelium mucosa are present but deep mucosal damage is 

absent. (c) pre-treated with DLELE (250 mg/kg). Moderate disruption of surface epithelium are 

present but deep mucosal damage is absent. There is edema and leucocytes infiltration of the 

submucosal layer. (d) pre-treated with DLELE (500 mg/kg). There is no disruption to the surface 

epithelium with no edema and no leuco. 

scavenge for the reactive oxygen species (super-oxide 

anions) and free radicals produced by ethanol. These 

reactive intermediates are potentially implicated in 

ulcerogenicity (Lewis and Hanson, 1991). 

It is generally known that the antioxidant activities of 

putative antioxidants involves various mechanisms, such 

as radical scavenging, decomposition of peroxides, 

binding of transition metal ion catalysts, prevention of 

chain initiation and of continued hydrogen abstraction 

(Diplock et al., 1998). Hence, the free radical scavenging 

capacity of an extract may serve as a significant indicator 

of its potential antioxidant activity. Increasing evidences 

have suggested that many age-related human diseases 

are the result of cellular damage caused by free radicals 

(Carr and Frei, 2000), Antioxidants have been shown to 

play an important role in preventing such diseases. For 


example, several cancer chemopreventive agents exhibit 

antioxidant activity through their ability to scavenge 

oxygen radicals (Ito et al., 1999). DLELE demonstrated to 

contain flavonoids, saponins, triterpenes, tannins and 

steroids (Zakaria, 2007). The interests in phenolic 

compounds, particularly flavonoids and tannins, have 

considerably increased in recent years because of their 

broad spectrum of chemical and diverse biological 

properties, which include the antioxidant effects (Larson, 

1988) and radical scavenging properties (Agrawal, 1989). 

Flavonoids have been associated with possible role in the 

prevention of several chronic diseases involving oxidative 

stress (Lee et al., 2003), as well as their protective effect 

against low-density lipoprotein (LDL) oxidation (Silva et 

al., 2000). Flavonoids have been reported to inhibit 

cytokine (inflammatory stimuli) release from RAW264.7


cells (Xagorari et al., 2002) and may modulate the 

increasing number of cellular processes involving redox 

reaction, including the regulation of tyrosine phosphatase 

activity (Gamet-Payrastre et al., 1999). 

The result of the present study also revealed protection 

of gastric mucosa and inhibition of leucocytes infiltration 

of gastric wall in rats pretreated with DLELE. DLELE 

have been shown to contain anti-inflammatory activity 

(Zakaria et al., 2006) and it is speculated that the 

gastroprotective effect exerted by this plant extract could 

be attributed to its anti-inflammatory activity. This antiinflammatory 

activity could also be a key factor in the 

prevention of gastric ulcer as reported by Swarnakar et 

al. (2005). Similarly, Abdulla et al. (2010) and Wasman et 

al. (2010) demonstrated that the reduction of neutrophil 

infiltration into ulcerated gastric tissue promotes the 

healing of gastric ulcers in rats. Mahmood et al. (2010) 

and Wasman et al. (2010) showed that oral administration 

of plant extract before ethanol administration 

significantly decreased neutrophil infiltration of gastric 

mucosa. Absolute alcohol would extensively damage the 

gastric mucosa leading to increased neutrophil infiltration 

into the gastric mucosa. Oxygenfree radicals derived 

from infiltrated neutrophils in ulcerated gastric tissues 

have inhibitory effect on gastric ulcers healing in rats 

(Suzuki et al., 1998). Neutrophils mediate lipid peroxidation 

through the production of superoxide anions 

(Zimmerman et al., 1997). Neutrophils are a major source 

of inflammatory mediators and can release potent 

reactive oxygen species such as superoxide, hydrogen 

peroxide and myeloperoxidase derived oxidants. These 

reactive oxygen species are highly cytotoxic and can 

induce tissue damage (Cheng and Koo, 2000). 

Furthermore, neutrophil accumulation in gastric mucosa 

has been shown to induce gastric ulceration (Abdulla et 

al., 2010; Wasman et al., 2010; Ketuly et al., 2011). 

Suppression of neutrophil infiltration during inflammation 

was found to enhance gastric ulcer healing (Mahmood et 

al., 2010). Studies have demonstrated the link between 

the anti-inflammatory and antioxidant activities of the 

plants. For example, nitric oxide (NO) is produced/ 

released under the action of inflammatory stimuli (that is, 

ROS) (Olszanecki et al., 2002). Inhibition of ROS leads to 

the reduction of NO production, which has been demonstrated 

to cause anti-inflammatory and antioxidant 

activities (Middleton et al., 2000). The free radical 

scavenging property may be one of the mechanisms by 

which these plants’ are effective in their ethnopharmacological 

uses against different ailments. 

In the present study, we observed flattening of the 

mucosal folds which suggests that gastroprotective effect 

of DLELE might be due to a decrease in gastric motility. It 

is reported that the changes in the gastric motility may 

play a role in the development and prevention of 

experimental gastric lesions (Abdulla et al., 2010; Ketuly 

et al., 2011). Relaxation of circular muscles may protect 

the gastric mucosa through flattening of the folds. This 

will increase the mucosal area exposed to necrotizing 

agents and reduce the volume of the gastric irritants on 

rugal crest (Mahmood et al., 2010; Wasman et al., 2010). 

Ethanol produces a marked contraction of the circular 

muscles of rat fundic strip. Such a contraction can lead to 

mucosal compression at the site of the greatest 

mechanical stress, at the crests of mucosal folds leading 

to necrosis and ulceration (Abdulla et al., 2010). 

Conclusion 

The study reveals DLELE could significantly protect the 

gastric mucosa against ethanol-induced injury. Such 

protection was ascertained grossly by increased gastric 

mucus production, decrease in the acidity of gastric 

content were significantly higher in treated groups 

compare to ulcer control group and also the reduction of 

ulcer areas in the gastric wall as well as histology by the 

reduction or inhibition of edema and leucocytes infiltration 

of submucosal layers. The data obtained confirm the 

traditional indications for this herb and present a new 

therapeutic option for the treatment of gastric ailments. 

The exact mechanism (s) underlying this anti-ulcerogenic 

effect remain unknown, but it seems that this extract 

contains pharmacologically active substances with potent 

antioxidant and anti-inflammatory activity which increase 

the mucus production and decrease the acidity of gastric 

content. 

ACKNOWLEDGEMENTS 

This study was financially supported by the University of 

Malaya through University Malaya Research Grant 2009 

(UMRG), RG102/09HTM and 600-RMI/ERGS 5/3 

(26/2011). 

REFERENCES 

Abdulla MA, Ahmed KAA, AL-Bayaty, FH, Masood, Y (2010). 

Gastroprotective effect of Phyllanthus niruri leaf extract against 

ethanol-induced gastric mucosal injury in rats. Afr. J. Pharm. 

Pharmacol., 4(5): 226-230. 

Agrawal PK (1989). Carbon-13 NMR of Flavonoids. New York: Elsevier 

Press, pp. 5-31. 

Al-Mofleh IA, Alhaider AA, Mossa JS, Al-Soohaibani MO, Rafatullah S 

(2007). Aqueous suspension of anise Pimpinella anisum protects rats 

against chemically induced gastric ulcers. World J. Gastroenterol., 

13: 1112-1118. 

Carr A, Frei B (2000). The role of natural antioxidants in preserving the 

biological activity of endothelium-derived nitric oxide. Free Rad. Biol. 

Med., 28: 1806-1814. 

Cheng CL, Koo MWL (2000). Effect of Centella asiatica on ethanol 

induced gastric mucosal lesions in rats. Life Sci., 67: 2647-2653. 

Chin WY (1992). A Guide to Medicinal Plants. Singapore Sci. Centre, p. 

24. 

Diplock AT, Charleux JL, Crozier-Willi G, Kok FJ, Rice-Evans C, 

Roberfroid M, Stahl W, Vina-Ribes J (1998). Functional food science 

and defence against reactive oxygen species. Br. J. Nutr., 80: S77- 

S112.

Gamet-Payrastre L, Manenti S, Gratacap MP, Tulliez J, Chap H, 

Payrastre B (1999). Flavonoids and the inhibition of PKC and PI 3- 

kinase. Gen. Pharmacol., 32: 279-286. 

Hiruma-Lima CA, Calvo TR, Rodrigues CM, Andrade FD, Vilegas W, 

Brito AR (2006). Antiulcerogenic activity of Alchornea castaneafolia: 

effects on somatostatin, gastrin and prostaglandin. J. 

Ethnopharmacol., 104: 215-224. 

Ito H, Miyake M, Nishitani E, Mori K, Hatano T, Okuda T, Konoshima T, 

Takasaki M, Kozuka M, Mukainaka M, Tokudo H, Nishino H, Yoshida 

T (1999). Antitumor promoting activity of polyphenols from Cowania 

mexicana and Coleogyne ramosissima. Cancer Lett., 143: 5-13. 

Khazaei M, Salehi H (2006). Protective effect of falcaria vulgaris extract 

on ethanol induced gastric ulcer in rat. Iran J. Pharmacol. Ther., 5: 

43-46. 

Ketuly KA, Abdulla MA, Hadi HA, Mariod AA, Abdel-Wahab SI (2011). 

Anti-ulcer activity of the 9alpha-bromo analogue of Beclomethasone 

dipropionate against ethanol-induced gastric mucosal injury in rats. J. 

Med. Plants Res., 5(4): 514-520. 

La Casa C, Villegas I, Alarcon de la Lastra C, Motilva V, Martin Calero 

MJ (2000). Evidence for protective and antioxidant properties in rutin, 

a natural flavones, against ethanol induced gastric lesions. J. 

Ethnopharmacol., 71: 45-53. 

Lai HY, Lim YY, Tan SP (2009). Antioxidative, tyrosinase inhibiting and 

antibacterial activities of leaf extract from medicinal ferns. Biosci. 

Biotechnol. Biochem., 73(6): 1362-1366. 

Larson RA (1988). The antioxidants of higher plants. Phytochem., 27: 

969-978. 

Lee KW, Kim YJ, Lee HJ, Lee CY (2003). Cocoa has more phenolic 

phytochemicals and a higher antioxidant capacity than teas and red 

wine. J. Agric. Food Chem., 51: 7292-7295. 

Lewis DA, Hanson PJ (1991). Antiulcer drugs of plant origin. Prog. Med. 

Chem., 28: 210-229. 

Ligumsky M, Sestieri M, Okon F, Ginsburg I (1995). Antioxidants inhibit 

ethanol-induced gastric injury in the rat. Role of manganese, glycin 

and carotene. Scand. J. Gastroenterol., 30: 854-860. 

Olszanecki R, Gêbska A, Kozlovski VI, Grylewski RJ (2002). Flavonoids 

and nitric oxide synthase. J. Physiol. Pharmacol., 53: 571-584. 

Mahmood AA, Mariod AA, Al-Bayaty F, Abdel-Wahab SI (2010). Antiulcerogenic 

activity of Gynura procumbens leaf extract against 

experimentally-induced gastric lesions in rats. J. Med. Plants Res., 

4(8): 685-691. 

Martin MJ, Marhuenda E, Perez-Guerrero C, Franco JM (1994). 

Antiulcer effect of naringin on gastric lesion induced by ethanol in 

rats. Pharmacology, 49: 144-150. 

Middleton Jr. E, Kandaswami C, Theoharides TC (2000). The effects of 

plant flavonoids on mammalian cells: implications for inflammation, 

heart disease, and cancer. Pharmacol. Rev., 52: 673-751. 

Mizui T, Sato H, Hirose F, Doteuchi M (1987). Effect of antiperoxidative 

drugs on gastric damage induced by ethanol in rats. Life Sci., 41: 

755-763. 

Raja DP, Manickam VS, De Britto AJ, Gopalakrishnan S, Ushioda T, 

Satoh M, Tanimura A, Fuchino H, Tanaka N (1995). Chemical and 

chemotaxonomical studies on Dicranopteris species. Chem. Pharm. 

Bull., 43: 1800-1803. 

Saurez J, Herreta MD, Marhuenda E (1996). Hesperidine and 

neohesperidine dihydrochalcone on different experimental models of 

induced gastric ulcer. Phytother. Res., 10: 616-618. 


Silva FA, Borges F, Guimaraes C, Lima JL, Matos C, Reis S (2000). 

Phenolic acids and derivatives: Studies on the relationship among 

structure, radical activity, and physicochemical parameters. J. Agric. 

Food Chem., 48: 2122-2126. 

Suzuki Y, Ishihara M, Ito M (1998). Anti-ulcer effects of antioxidants, 

quercetin, �-tocopherol, nifedipine and tetracycline in rats. Jpn. J. 

Pharmacol., 78: 435-441. 

Swarnakar S, Ganguly K, Kundu P, Banerjee A, Maity P, Sharma AV 

(2005). Curcumin regulates expression and activity of matrix 

metalloprotinases 9 and 2 during prevention and healing of 

indomethacin-induced gastric ulcer. J. Biol. Chem., 280: 9409-9415. 

Tanaka JCA, Da Silva CC, De Oliveira AJB, Nakamura CV, Dias Filho 

BP (2006) Antibacterial activity of indole alkaloids from Aspidosperma 

ramiflorum. Braz. J. Med. Biol. Res., 39: 387-391 

Trivedi NP, Rawal UM (2001). Hepatoprotective and antioxidant 

property of A. paniculata Nees in BHC induced liver damage in mice. 

Indian J. Exp. Biol., 39: 41-46. 

Vasuda SM (1999). Economic importance of pteridophytes. Indian Fern. 

J. 16: 130-152. 

Wasman SQ, Mahmood AA, Salehhuddin H, Zahra AA, Salmah I 

(2010). Cytoprotective activities of Polygonum minus aqueous leaf 

extract on ethanol-induced gastric ulcer in rats. J. Med. Plants Res., 

4(24): 2658-2665. 

Xagorari A, Roussos C, Papapetrpoulos A (2002). Inhibition of 

LPSstimulated pathways in macrophages by the flavonoid luteolin. 

Br. J. Pharmacol., 136: 1058-1064. 

Zakaria ZA (2007). Free radical scavenging activity of some plants 

available in Malaysia. Iran J. Pharmacol. Ther., 6: 87-91. 

Zakaria ZA, Abdul Ghani ZDF, Raden Mohd Nor RNS, Hanan Kumar G, 

Sulaiman MR, Fatimah CA (2006). Antinociceptive and 

antiinflammatory activities of Dicranopteris linearis leaves chloroform 

extract in experimental animals. Yakugaku Zasshi, 126: 1197-1203. 

Zakaria ZA, Ghani ZD, Nor RN, Gopalan HK, Sulaiman MR, Jais AM, 

Somchit MN, Kader AA, Ripin J (2008). Antinociceptive, antiinflammatory, 

and antipyretic properties of an aqueous extract of 

Dicranopteris linearis leaves in experimental animal models. J. Nat. 

Med., 62(2):179-87. 

Zakaria ZA, Mohamed AM, Mohd. Jamil NS, Rofiee MS, Somchit MN, 

Zuraini A, Arifah AK, Sulaiman MR (2011). In vitro cytotoxic and 

antioxidant properties of the aqueous, chloroform and methanol 

extracts of Dicranopteris linearis leaves. Afr. J. Biotechnol., 10(2): 

273-282. 

Zayachkivska OS, Konturek SJ, Drozdowicz D, Konturek PC, 

Brzozowski T, Ghegotsky MR (2005). Gastroprotective effects of 

flavonoids in plant extracts. J. Physiol. Pharmacol., 56: 219-231. 

Zimmerman JJ, Ciesielski W, Lewandoski J (1997). Neutrophilmediated 

phospholipids peroxidation assessed by gas 

chromatography-mass spectroscopy. Am. J. Physiol., 273: 653-661.



DOI: 10.5897/SRE11.1483 

ISSN 1992-2248 © 2012 Academic Journals 


Introducing slag powder as drag reduction agent in 

pipeline: An experimental approach 

Hayder A. Abdulbari*, Siti Nuraffini Bt Kamarulizam, Rosli M. Y. and Arun Gupta 

Faculty of Chemical and Natural Resources Engineering, University Malaysia Pahang, Lebuhraya Tun Razak, 26300 

Kuantan, Pahang Darul Makmur, Malaysia. 

Accepted 20 September, 2011 

Main by product in ore smelting from tin production in Malaysia has become a trigger for this 

investigation. Slag waste can be categorized as suspended solid. Utilization of this waste in fluid 

transportation can reduce the pressure drop in pipelines. Experimental works had conducted in order to 

test slag waste in a closed loop of turbulence water flowing system with water and fuel as the transport 

liquid. The procedures start by pumping liquid suspended solid combination from reservoir tank with 

varies flow rates into two different pipe diameters (0.0127 m ID and 0.038 m D.I). The types of pipe used 

are PVC pipe. The testing length of this flow system is 2.0 m. The pressure drop and drag reduction 

were measured in varied addition concentration. The results have show percentage drag reduction 

(Dr%) is over 60% in certain range and condition. It is proved that slag is a potential DRA. 

Key words: Suspended solid, turbulent flow, drag reducing agent, pipeline system. 

INTRODUCTION 

Drag reduction in turbulent flow is a very important 

subject in technologies utilization and significant point of 

interest. As known, drag reduction can be achieved by 

using several numbers of additives that have been widely 

studied such as high molecular weight polymers (Roy 

and Larson, 2005; Al-Sarkhi, 2010, Mowla and Naderi, 

2006; Shetty and Solomon, 2009; Parimal et al., 2008; 

Janosi et al., 2004; Dubief et al., 2004) surfactants study 

by Lu et al. (1998), Bari and Yunus (2009), Bari et al. 

(2008) and Suali et al. (2010) and suspended solid 

investigation have been determined by Roy and Larson 

(2005) and Dyer et al. (2004). Drag reduction is defined 

as addition of several ppm concentrations to accelerate 

radically in fluid transportation (Brostow et al., 2006; Suali 

et al., 2010; Dubief et al., 2004; Kim et al., 2000; Parimal 

et al., 2008) and many more. However, there are few 

studies regarding suspended solid. Mechanism involved 

is yet still a hesitation. Many researchers investigated the 

idea of particles inside liquid flow channel agreed in one 

point, turbulent characteristic are changed in present 

*Corresponding author. E-mail: hayder.bari@gmail.com. Tel: 

0060123495130. 

of suspended solid (Rashidi et al., 1990; Roy and Larson, 

2005; Dyer et al., 2004; Filipson et al., 1977). There are 

fewer studies devoted to the effects of particle additives 

on the mechanisms of instability and transition to 

turbulence in free shear flows. The flow visualizations 

reported by Filipsson et al. (1977) represent one of the 

few available experiments on this subject. In this study, 

the authors presented results for a jet flow of viscoelastic 

(Polyox WSR-301), fibre suspension (chrysotile fibres) 

and Newtonian (water) fluids at high Reynolds numbers. 

Rashidi et al. (1990) have determined predominant 

effects of particle size, density and concentration. Their 

results point out that particle density has minor effects 

rather than particle size in effects of drag reduction. 

Presently, the usage of the suspended solid (insoluble 

in liquid media) as drag reducing agent has broaden 

accesses for enthusiastic researcher to uncover the 

accessibility of this insoluble condition in the drag 

reduction phenomena (Toonder, 1997; Mowla and 

Naderi, 2006). The aim of this study is to test the 

efficiency of slag particle as drag reducer agent on 

transport of fuel inside pipes. Two different internal pipe 

diameters were used with four different concentrations in 

the purpose to investigate the concentration effect. The 

efficiency of suspended solid was tested using diesel.

Instrument 

P1 – P5 Pressure sensor for 0.0381 m D.I 

P6 – P10 Pressure sensor for 0.0254 m D.I 

Valve 

Centrifugal pump 

Figure 1. Schematic diagram of the flow system. 


Liquid circulation system 

Liquid circulation system was built to test the effects of pipe 

diameter, pipe length, fluid velocity and concentration on pressure 

drop; hence to investigate the effects influence %Dr. Figure 1 

shows a schematic diagram of a build up liquid circulation system 

used in the present investigation. Generally, this system consists of 

reservoir tank, pipes, valves, pumps, flow meter and pressure 

sensors. The reservoir tank was supported with an exit pipe 

connected to centrifugal pumps. The solution flows from reservoir 

tank directly into testing pipes before flowing back into the reservoir 

tank. Two visible PVC pipes connected to galvanized iron pipes of 

various inside diameters 0.0127 and 0.0381 m ID were used in 

constructing the flow system. A complete closed loop piping system 

Abdulbari et al. 1769 

was built. Flow starts from the reservoir tank through the pump 

reaching a split connected with three different pipes diameter with 

testing section. For measuring flow rate inside the pipe, Burkert 

attachable flow meter has been used. Power supply for this 

detector is 12 to 30 V and can measure flow pressure between 140 

to 230 psi with fluid viscosity les then 200 cP. Material used to be 

immersed into fluid is ceramic. The pump brand for water circulation 

is Grundfos CH8-40. The energy usage of this pump is 1.02 kW 

and pump force is 415 V. On the other hand, hydrocarbon pump 

brand is DAB K-Series which is transfer pump for small flow 

operations. This centrifugal pump is designed with technopoymer 

impeller for domestic usage. The energy required for this pump is 

0.75 kW and pump force is 240 V. The testing sections were 0.5, 

1.0, 1.5 and 2.0 m long. First, testing point was located about 50 

times of pipe diameter to ensure that the turbulent flows are fully 

developed before the testing process run. Five sets of build up


(a) (b) 

Figure 2. Grinded slag waste particle for 200 μm: (a) SEM picture of 9x mag and (b) Scaled up picture of a particle 

(33x mag). 

Table 1. Physical properties of diesel. 

Hydrocarbon properties at 27°C 

Viscosity (μ diesel at 27°C ) 1456 cP 

Density (ρ diesel at 27°C ) 853.2 kg/m 3 

Table 2. Physical properties of water. 

Water properties at 27°C 

Viscosity (μ water at 27°C ) 937.3 cP 

Density (ρ waterat 27°C ) 996.59 kg/m 3 

pressure sensors. 

The pressure sensor is a build up pressure transmitter and 

transducer made by silicon as the base materials. The membrane 

detector is made by 316 L stainless steel diaphragm. The optimum 

range for pressure is 0 to 60 bar, temperature range between -10 

until 60°C and be detected in absolute gauge. The accuracy for this 

transmitter is 0.1% fault. These sensors are integrated construction 

specially design for rigid and robust flow in industry. 

Materials investigated (slag waste particle) 

Slag waste is obtained from Malaysia Smelting Corporation at 

Butterworth, Penang. The slag waste was dried by the oven 

overnight at a temperature of 100°C. Once dry, the suspended solid 

is graded into fine particle by using grinder then sample was sieve 

using a screen into 200 μm in size (Figure 2). The density of this 

new material is 1400 kg/m 3 . 

Transported liquid 

The transported liquid used in the present investigation was diesel 

fuel obtained from Shell. However, for comparison, pure water was 

used. The physical properties of diesel fuel and water are shown in 

Tables 1 and 2. 

Experimental procedure 

All the experiments were carried in a constructed liquid circulation 

system, testing different variables which are: 

i) Suspended solid concentration (50, 100, 200 and 400 ppm). 

ii) Pipe diameter (0.0127 and 0.0381 m D.I). 

iii) Pipe length (0.5, 1.0, 1.5 and 2.0 m). 

iv) Solution flow rates.


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 

and pressure drop analysis for Re = 105000. 

The experimental procedure starts by testing every additive 

concentration and pipe diameter, the operation begins when the 

pump starts delivering the solution through section of pipe length. 

The solution flow rate is fixed at the certain value by controlling it 

from the bypass section. Pressure readings are taken to this flow 

rate. By changing the solution flow rate to another fixed point, 

pressure readings are taken again until finishing desired values of 

flow rates. This procedure is repeated for each suspended solid 

concentrations to test its effect on the drag reduction operation. 

From the flow rate obtain, Reynolds number calculated based on 

formulae, which represent velocity of flow depending on pipe 

diameter and pipe length. 

Velocity and Reynolds number calculations 

The average velocity (V) and Reynolds number (Re) were 

calculated using the solution volumetric flow rate readings (Q), 

density (ρ), viscosity (μ) and pipe diameter (D) for each run as 

follows: 

�V. . D 

Re� 

� 

Where: 

ρ is density of the fluid (kg/m 3 ), 

µ is viscosity of the fluid (Pa. s), 

D is diameter of internal pipe (m), 

V is velocity of fluid (m/h), 

Where: = 

(1) 

Where: = 

After obtaining Reynolds number, percentage drag reduction (%Dr) 

determined to plot graft in order to see the patent of drag reduction. 

Percentage drag reduction calculations 

Pressure drop readings through testing sections before and after 

drag reducer addition were needed to calculate the percentage 

drag reduction %Dr as follows (Virk, 1967): 

Pb 

Pa 

Pb 

% Dr�� 

� 

Where: 

� � 

∆Pb = Pressure drop before addition of DRA. 

∆Pa = Pressure drop after addition of DRA. 

RESULTS 

The analysis in Figures 3 and 4 shows slag particle 

performance on drag reduction as a function of particle 

concentration in hydrocarbon liquid and water and fluid 

velocity represented by Re which is from 50 to 400 ppm 

in the range of Re equal to 65000 to 140000 for fuel and 

65000 to 105000 for water transportation. Scale down for 

(2)


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 

pressure drop analysis for Re = 105000. 

Figure 5. Effect of pipe diameter flowing through 2 m pipe length in fuel and 400 ppm concentration addition and pressure drop 

data for pipe diameter of 0.0127 and 0.0381 m D.I. 

one Re showed different pressure drop from raw data 

obtained as seen in Figures 3 and 4 for Re = 14000 for 

fuel and 105000 for water. The analysis in Figures 5 and 

6 shows slag particle performance on drag reduction as a 

function of pipe diameter in hydrocarbon liquid and water 

and fluid velocity represented by Re which is for 400 ppm 

in the range of Re equals 45000 to 145000 for fuel and 

Re equals 32000 to 95000 for water. Scale down for one 

Re, the different of pressure drop in two different pipe 

diameters. For hydrocarbon, Re for raw pressure drop 

data taken at 105000 for 0.0127 m D.I and 145000 for 

0.0381 m D.I. For water, Re for raw pressure drop data


Figure 6. Effect of pipe diameter flowing through 2 m pipe length in water and 400 ppm concentration addition and pressure 

drop data for pipe diameter of 0.0127 and 0.0381 m D.I. 

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 

data for Re = 139000. 

taken at 70000 for 0.0127 m D.I and 95000 for 0.0381 m 

D.I. The analysis in Figures 7 and 8 shows slag particle 

performance on drag reduction as a function of pipe 

length in hydrocarbon liquid and water and fluid velocity 

represented by Re which is for 400 ppm in the range of 

Re equal to 98000 to 145000 for fuel and Re equal 65000 

to 105000 for water. Scale down for one Re, the different 

of pressure drop in four different testing sections which


Figure 8. Effect of pipe length in 400 ppm addition concentration flowing in water with pipe diameter 0.0381 m D.I and 

pressure drop data for Re = 100000. 

are 0.5, 1.0, 1.5 and 2.0 m. For hydrocarbon, Re taken at 

point equal to 139000 and for water, Re taken at point 

equal to 100000. 

DISCUSSION 

Note that the profile pattern of each concentration and 

fluid velocity is similar but varies in its value. Both result 

shows, either in fuel or water, drag reduction is increased 

by increasing of particle addition concentration. These 

data complied with many other researchers such as 

Rashidi et al. (1990), Roy and Larson (2005), Dyer et al. 

(2004) and Filipson et al. (1977). There are two result 

which shows drag reducer performance towards each 

solvent. These results showed that the optimum 

performances of suspended solid additive are limited to 

the degree of turbulence. The increment of fluid velocity 

in flow will establish a degree of turbulences in its own 

fluid range. However, there are two explanations that can 

be made for these two results. For Figure 3, drag 

reduction efficiency are decreased by increasing 

Reynolds number. However, drag reduction are still 

recorded in the flow. For different type of solvent, 

efficiency for same drag reducer are different. Due to 

lighter fuel density and viscosity compare to water, slag 

particle seem too ‘heavy’ to be pushed; so that, along 

pipe flow, the particle are deposited at the bottom of pipe 

wall little by little and result to few particle to stay flow 

with fluid. These phenomena have been explained by 

Dyer et al. (2004). Figure 4 shows higher fluid velocity 

gives higher degree of turbulent which will provide more 

suitable environment or possibility for the drag reduction 

mechanism to perform. Further increase can happen until 

it comes to the limitation. Over its degree of turbulent will 

cause the reduction static due to decrement of additive 

efficiency. The reason for these phenomena is because 

of the impairing ratio of additive concentration and degree 

of turbulence. 

Figures 5 and 6 shows the same pattern of drag 

reduction in function of additive concentration and fluid 

velocity. However, the value in 0.0127 m D.I is obviously 

smaller compared to 0.0381 m D.I. Increasing the pipe 

diameter means increasing the velocity inside the pipe in 

range of same flow rate resulting in increment of 

turbulence. The presence of eddies means that the local 

velocity is not the same as the bulk velocity and that 

there are components of velocity in all directions. The 

result complied with Bari et al. (2008). In larger pipe, the 

energy in the pipe is larger than smaller pipe due to fluid 

quantity itself. Higher energy is maintained inside large 

pipe rather than small pipe because of less different of 

local and bulk velocity hence decrease shear stress 

formation. However, in smaller pipes, high pressures are 

obtained because of fluid flow from large area to small 

area. When pressure is high, shear stress between local 

velocity and bulk velocity is significantly, resulting to more 

amount of energy absorbed to form eddies. However, 

due to higher volume of fluid, higher degree of turbulence 

are obtained in larger pipes because of higher Reynolds 

number resulting to higher possible collisions between 

eddies and more space of DRA to take action. These 

collisions provide extra number of eddies absorbing 

energy from the main flow to complete their shape. By

addition of DRA, eddies is bursting and pressure can be 

maintained along the pipes. However, pressure are not at 

the same maintaining state in order to burst eddies along 

the pipes. Also include friction of particle itself with pipe 

walls will reduce the efficiency. This can be proofed by 

Figures 5 and 6. Both figure shows by increasing pipe 

length, most of %DR is decreased slightly after 

maintaining the reduction for a while. For additional 

information, particle diameter effect towards slag particle 

flexibility to act as drag reducer. In order to be a good 

drag reducer, basic criteria such as flexibility and surface 

roughness should be fulfilled (Singh, 1990). However, 

fluctuation in drag reduction reading is due to high 

turbulent intensities which mean that the flow is very 

unstable. This phenomena happen in all variables 

(concentrations, time, pipe length and pipe diameter) as 

can be seen in Figures 3, 4, 5, 6, 7 and 8 for both solvent 

conditions. 

From Figures 7 and 8, drag reduction have occur in first 

section which means higher drag reduction and the 

effectiveness lesser within the length due to less 

momentum inside these particles. As known, slag particle 

cannot be degraded so that when particles are ‘pushed’ 

by instable turbulent flow; it then increased the 

momentum back and performs drag reduction again. 

These processes will be repeated along the pipe as long 

as these particle remains inside the flow and not be 

degraded by any condition. For the industrial activities, 

longer pipes and bigger diameters are vital in long 

distance transportation. The patterns that drag reduction 

have proved can be considered as potential DRA. Due to 

very stable particles, the drag reduction will give up and 

down result as explained earlier. However, these 

particles are maintained to give effect on drag reduction 

along pipe length without addition of extra DRA. Bigger 

pipe also give advantages to this DRA since more space 

for DRA mechanism to perform and give less turbulent 

intensities hence increased drag reduction efficiency. As 

proved in these graphs data and present results, there 

are some weaknesses in using present DRA. Time 

consumption is needed in order to separate DRA with 

transported solvent. The separation begins by 

sedimentation of DRA to bottom of tank and transported 

solvent can be flow gravitational or by pump to other 

section for further use. However, because this particle is 

inert and stable which do not react with transported liquid, 

letting DRA immersed for a while is not a big problem. 

These present DRA also give certain advantages 

compared to polymeric DRA, surfactant DRA and fibers 

DRA. Compared to other DRA, there were no reinsertions 

of DRA once added into transported liquid. This will 

reduce operation cost. By using present DRA, the 

characteristic and properties of transported liquid will not 

ever be disturbed. Compare to polymer, degradation is a 

major problem since polymer chain can broke due to 

mechanical force (pump rotation) and biodegradable over 

time. For surfactant, micelle formation due to shear stress 


will reduce drag reduction efficiency. 

In hydrocarbon, surfactant can trap water to move 

along and will give emulsion. This emulsion is a great 

problem since to separate required much more time and 

temperature control. Fibre which categorized as 

suspended solid also is not very suitable in transporting 

hydrocarbon since it can change the hydrocarbon 

characteristic. 

Conclusion 

It is concluded that slag waste particle is applicable as 

suspended solid DRA since it is economic, inert (do not 

react with transported fluid), effective in water and diesel, 

potentially in refinery products. Several effects have been 

investigated in order to this new economic solver and 

how the drag reducing work. It had proved that slag 

particle will increase %Dr by increasing fluid velocity and 

concentration and pipe diameter. There is several 

limitation of using this suspended solid. Suspended solid 

can be effective in certain range degree of turbulent but 

will give sedimentation at the base of storage tank. As for 

recommendation, stirrer can be installed inside tank to 

ensure particles are evenly distributed. This suspended 

solid is not harmful to living organism and environment 

since there is no hazardous chemical that have been 

used. 


Deepest gratitude is expressed to Universiti Malaysia 

Pahang and Faculty of Chemical and Natural Resources 

Engineering for providing the grant and facilities to 

support this research work. Appreciation extended to the 

supervisor, co-supervisor and laboratory staff for their 

guidance, advice, support and encouragements. 

REFERENCES 

Abdul Bari HA, Mohd RY (2009). Sodium Stearate as Drag Reducing 

Agent in Non-Aqueous Media. Int. J. Chem. Tech., 1(1): 11-18. 

Al-Sarkhi A (2010). Drag reduction with polymers in gas-liquid/ liquidliquid 

flows in pipes: A Literature Review. J. Nat. Gas Sci. Eng., 2: 

41-48. 

Bari HAA, Suali E, Hassan Z (2008). Glycolic acid ethoxylate lauryl 

ether performances as drag reducing agent in aqueos media flow in 

pipelines. J. Appl. Sci., 8: 4410-4415. 

Brostow W, Haley EHL, Reddy T, Singh RP, White L (2006). Lowering 

mechanical degradation of drag reducers in turbulent flow. J. Mater. 

Res., 22(1): 56-60. 

Dubief Y, White CM, Terrapon VE, Shaqfeh ESG, Moin P, Lele SK 

(2004). On the coherent drag-reducing and turbulence-enhancing 

behavior of polymers in wall flows. J. Fluid Mech., 514: 271-280. 

Dyer KR, Christie MC, Manning AJ (2004). The effects of suspended 

sediment on turbulence within an estuarine tirbudity maximum. J. Est. 

Coast Shelf Sci., 59: 237-248. 

Filipsson LGR, Torgny Lagerstedt JH, Bark FH (1977). A note on the 

analogous behavior of turbulent jets of dilute surfactant solutions and 

fibre suspensions. J. non-Newt Fluid Mech., 3: 97-103.


Janosi IM, Jan D, Szabó KG, Tel T (2004). Turbulent drag reduction in 

dam-break flows. Exp. Fluids., 37: 219-229. 

Kim CA, Kim JT, Lee K, Choi HJ, Jhon MS (2000). Mechanical 

degradation of dilute polymer solutions under turbulent flow, J. Pol., 

41: 7611-7615. 

Lu B, Li X, Scriven le, Davis HT, Talmon Y, Zakin JL (1998). Effect of 

Chemical Structure on Viscoelasticity and Extensional Viscosity of 

Drag-Reducing Cationic Surfactant Solutions. J. Chem. Struct. 

Visco, Langmuir., 14(1): 8-16. 

Mowla D, Naderi A (2006). Experimental study of drag reduction by a 

polymeric additive in slug two-phase flow of crude oil and air in 

horizontal pipes. Chem. Eng. Sci., 61: 1549-1554. 

Parimal PM, Cheolho K, Alvaro AOM (2008). The Performance of Drag 

Reducing Agent in Multiphase Flow Conditions at High Pressure; 

Positive and Negatives Effects. Proc. Of 7th Int. Pipeline Conf., 

Calgary, Alberta, Canada. 

Rashidi M, Hetsroni G, Banerjee S (1990). Particle-turbulence 

interaction in a boundary layer. Int. J. Mult. Flow, 16(6): 935-949. 

Roy A, Larson RG (2005). A mean flow model for polymer and fiber 

turbulent drag reduction. App. Rheol. J., 15: 370-389. 

Shetty AM, Solomon MJ (2009). Aggregation in dilute solutions of high 

molar mass poly(ethylene) oxide and its effect on polymer turbulent 

drag reduction. Polym. J., 50: 261-270. 

Singh RD (1990). Encyclopedia of Fluid Dynamics, Vol. 9, Gulf 

Publishing Co., Houston, Texas, Chapt. 14: 425-480. 

Suali E, Abdul Bari HA, Hassan Z, Rahman MM (2010). The Study of 

Gycolic Acid Ethoxylate 4-nonylphenyl Ether on Drag Reduction. J. 

Appl. Sci., 10(21): 2683-2687. 

Toonder JMJD, Hulsen MA, Kuiken GDC, Nieuwstadt FTM (1997). Drag 

reduction by polymer additives in a turbulent pipe flow: Laboratory 

and numerical experiments. J. Fluid Mech., 337: 193-231. 

Virk PS, Merrill EW, Mickley HS, Smith KA (1967). The Toms 

phenomenon: Turbulent pipe flow of dilute polymer solutions. J. Fluid 

Mech., 30: 305-328.



DOI: 10.5897/SRE11.1759 



Available transfer capability and least square method 

Mojgan Hojabri 1 *, Mohammadsoroush Soheilirad 2 and Mahdi Hedayati 2 

1 Faculty of Electrical and Electronics Engineering, University Malaysia Pahang (UMP), 26600 Pekan, Pahang, Malaysia. 

2 Department of Electrical and Electronic Engineering, University Putra Malaysia (UPM), 43300Serdang, Selangor, 

Malaysia. 

Accepted 17 February, 2012 

In deregulated power industries, accurate and fast calculation of Available Transfer Capability (ATC) for 

seller in the electricity market is required. In this paper, deterministic ATC will be calculated using 

algebraic equation and linear optimization by Least Square (LSQR). Probabilistic ATC is also calculated 

by considering time varying load and load margin. The proposed method will be tested on IEEE 30 bus 

system. Deterministic ATC results will be compared with Benders, OPF and HCACO and probabilistic 

ATC results also will be compared with GA and HCACO. 

Key words: Available transfer capability (ATC), power system planning, LSQR. 

INTRODUCTION 

Competition in the electric power industry between sellers 

and buyers in power marketing poses new challenges for 

power system companies and researchers to find the 

best strategy for having beneficial energy trading. The 

technical challenges are related to the generation and 

transmission. Managing an effective operation can be 

provided by minimizing the operational cost, maximizing 

utilization of generators and transmission lines. However, 

power transmission systems are limited by the power 

transfer. The Available Transfer Capability (ATC) is 

required to be reported on the Open Access same–time 

Information System (OASIS) to inform all energy market 

participants of the maximum power transfer capability in 

power systems. Therefore to improve the power system 

efficiency and economy, a good strategy for calculating 

ATC is required. Moreover the strategy can be used to 

predict ATC for the future transmission enhancement in 

power system planning. 

Available transfer capability calculation is important for 

electric power companies and energy buyers. ATC 

calculation not only determines the energy transfer 

bounds but it also determines the reliability of the system 

in unsecured situations. Based on the definition of ATC 

by NERC in 1996, several researches have been done 

*Corresponding author. E-mail: mojgan.hojabri@gmail.com. 

on deterministic and probabilistic ATC calculations. Their 

objective was to find a fast and accurate method. From 

these researches, the speed of the deterministic ATC 

methods is better than probabilistic ATC calculation 

methods since in probabilistic ATC calculation 

uncertainties are considered. 

Probabilistic or stochastic power flow methods are used 

to accommodate the random nature of the operational 

load and generation data. Three important methods for 

stochastic power flow techniques are Monte Carlo, 

Convolution and Stochastic Algebraic. The Monte Carlo 

is a famous method to solve stochastic power flow 

problem (Huang and Yan, 2002; Yajing et al., 2005; Gao 

et al., 2006; Anselmo et al., 2007). This method by 

repeated trials of the deterministic ATC, calculates the 

probability distributions of the nodal powers, line flow and 

losses. The big disadvantage of Monte Carlo is 

computational burden. 

Convolution method calculated the impact of the 

uncertainty of load data to uncertainty of bus voltage and 

line power flow by (Borkowska, 1974). Some problem of 

this research is nonlinear relation between node leads 

and branch flows, and proper balance of generation and 

loads. In Stochastic power flow problem used (Dopazo et 

al., 1975), they assumed normally distributed generator 

for bus variables P and V. Then, they calculated power 

flows by using classical methods. Normality distributed 

complex random variables is the difficulty of this research


as described in (Sauer, 1977; Feller, 1971). 

Stochastic Algebraic method (Stahlhut et al., 2005; 

Jonathan, 2007) is one of the earliest probabilistic ATC 

calculation methods. Using full AC load flow and linear 

algebraic equations make this method simple. To 

overcome the problem of using linear method for 

nonlinear problem, Krylov subspace method is used in 

this study as a power iterative mathematical method. In 

Stochastic Algebraic calculation, ATC is only calculated 

for bilateral transactions. The usefulness of a bilateral 

transaction is ATC can easily be calculated between two 

buses, where the transaction power enters and leaves 

the network without considering various margins like 

transmission reliability margin, capacity benefit margin 

etc. for ATC evaluation. Therefore ATC is approximately 

equal to Total Transfer Capability (TTC), which is a key 

component for ATC assessment. In this paper, ATC is 

determined for multilateral transaction based on linear 

optimization using LSQR method. The impact of other 

lines, generators and loads on power transfer could be 

taken into account. Then the ATC computation will be 

more realistic. Another benefit of this method is by using 

linear programming, which makes the ATC computations 

simple. Moreover, the nonlinear behavior of ATC 

equations are considered by using one of the best 

iteration methods called Krylov subspace method. It is a 

robust method that could handle the nonsymmetrical and 

definite program (Ioannis, 2007). 

METHODOLOGY 

ATC definition 

In this paper, ATC are defined by linear optimization. To maximize 

the ATC (Equation 1), the objective function for the calculation of 

ATC is formulated as (Gnanadass and Ajjarapu, 2008): 

Where� P and 

gi � P are total power generated in the 

gj 

sending and receiving area. And� Pli and� P are total 

lj 

power used in the sending and receiving area. 

The objective function measures the power exchange between the 

sending and receiving areas. The constraints involved include, 

a) Equality power balance constraint. Mathematically, each lossless 

bilateral transaction between the sending and receiving bus i must 

satisfy the power balance relationship. 

(1) 

(2) 

(3) 

For multilateral transactions, this equation is extended to: 

Where is the total number of transactions. 

b) Inequality constraints on real power generation and utilization of 

both the sending and receiving area. 

Where and are the values of the real power 

generation and utilization of load flow in the sending and receiving 

areas, and are the maximum of real power 

generation and utilization in the sending and receiving areas. 

c) Inequality constraints on power rating and voltage limitations. 

With use of algebraic equations based load flow, margins for ATC 

calculation from bus i to bus j are represented in Equations (7and 8) 

and Equations (10 and 11). For thermal limitations the equations 

are, 

Where is determined as in Equation 8. 

Where and are bus voltage of the sending and receiving 

areas. And is the reactance between bus i and bus j. For 

voltage limitations, 

(4) 

(5) 

(6) 

(7) 

(8) 

(9) 

(10) 

(11) 

Where and are calculated as (Jonathan, 

2007): 

(12)

(13) 

Where is susceptance and represents a 

diagonal matrix whose elements are (for each transmission 

line), L is the incident matrix, PF is the power factor, and E11, E12, 

E21 and E22 are the sub matrixes of inverse Jacobian matrix. This 

can be achieved by steps descrided later. Reactive power (Q) 

constraints must be considered as active power constraints in 

Equations 3 to 6. 

Due to nonlinear behavior of power systems, linear 

approximation and can yield 

errors in the value of the ATC. In order to get a more precise ATC, 

an efficient iterative approach must be used. One of the most 

powerful tools for solving large and sparse systems of linear 

algebraic equations is a class of iterative methods called Krylov 

subspace methods. The significant advantages are low memory 

requirements and good approximation properties. To determine the 

ATC value for multilateral transactions the sum of ATC in Equation 

14 must be considered, 

Where k is the total number of transactions. 

(14) 

Krylov subspace methods form the most important class of iterative 

solution method. Approximation for the iterative solution of the 

linear problem for large, sparse and nonsymmetrical Amatrices, 

started more than 30 years ago (Adam, 1996). The 

approach was to minimize the residual r in the 

formulation . This led to techniques like, 

Biconjugate Gradients (BiCG), Biconjugate Gradients Stabilized 

(BICBSTAB), Conjugate Gradients Squared (CGS), Generalized 

Minimal Residual (GMRES), Least Square (LSQR), Minimal 

Residual (MINRES), Quasi-Minimal Residual (QMR) and Symmetric 

LQ (SYMMLQ). 

The solution strategy will depend on the nature of the problem to 

be solved which can be best characterized by the spectrum (the 

totality of the eigenvalues) of the system matrix A. The best and 

fastest convergence is obtained, in descending order, for A being: 

(a) Symmetrical (all eigenvalues are real) and definite, 

(b) Symmetric indefinite, 

(c) Nonsymmetrical (complex eigenvalues may exist in conjugate 

pairs) and definite real, and 

(d) Nonsymmetrical general 

However MINRES, CG and SYMMLQ can solve symmetrical and 

indefinite linear system whereas BICGSTAB, LSQR, QMR and 

GMRES are more suitable to handle nonsymmetrical and definite 

linear problems (Ioannis, 2007). The ATC margins equations can 

be represented in the general form: 

(15) 

Where represents vector form (number of branches) from 

Equations 7 and 8 and also vector form (number of buses) 

of Equations (10 and 11). With iteration step k, Equation 15 gives 

the residual r k. 

And the linearized form is: 

Hojabri et al. 1779 

(16) 

(17) 

Where A represents or 

in diagonal matrix form (number of branches) 

x (number of branches) or (number of buses) x (number of buses), 

and b gives or in vector 

form (number of branches) and or 

in vector form (number of buses) while the inequalities (7, 8, 10 and 

11) can be rewritten as in Equations 18-21. In this case, the nature 

of A is nonsymmetrical and definite. However, all of the Krylov 

subspace methods can be used for ATC computation but 

BICGSTAB, LSQR, QMR and GMRES are more suitable to handle 

this case. 

(18) 

(19) 

(20) 

(21) 

In this paper, LSQR is used for ATC computation. Numerically, 

LSQR is more reliable in various circumstances than the other 

Krylov subspace methods (Christopher and Michael, 1982). Small 

residual and using the standard QR factorization are other 

advantages of LSQR method (Golub and Kahan, 1965). 

Statistical analysis 

Statistical moments are used to provide some sort of measure for a 

probability distribution of ATC. The most important and useful 

moment is the center of a distribution of X. This center is called the 

mean, and is usually denoted as the , or the expectation of 

random variable ATC (Daniel and Ralph, 1996), 

(22) 

The mean is just one measure that a probability distribution has. 

The variance is another popular statistical measure of a probability 

distribution. The variance is a measure of the spread of the 

distribution. The variance is typically symbolized as 

or . The square root of variance is called standard


deviation . 

Figure 1. IEEE 30 Bus system. 

Table 1. Basic statistics for expected ATC (IEEE 30 bus system). 

Indices 

T1 

Line Outage 

T2 T3 

Time Varying Load 

T1 T2 T3 

Mean 103.6 96.76 44.437 106.81 102.93 48.034 

Standard Deviation 8.36 12.72 5.592 2.61 2.41 0.579 

Skewness -1.9 -2.6 -1.84 0.06 0.06 0.07 

Kurtosis 4.06 8.82 4.3 -1.53 -1.53 -1.52 

(23) 

(24) 

Where is the mean of ATC and N is the number of data. Two 

additional measures are used with the mean and standard deviation 

to help describe a probability distribution. These additional 

measures are skewness and kurtosis. The skewness is defined as: 

(25) 

When is the third moment of the mean of ATC and is the 

standard deviation. The measure of skewness is often useful for 

nonsymmetrical distribution. If the tail of the distribution is longer on 

the right, the skewness is a positive number. The skewness of a 

normal distribution is zero. The kurtosis of ATC is defined as in 

Equation 26. The kurtosis is a measure of peakedness of a 

distribution. A distribution that has a high kurtosis can range from -2 

to . The kurtosis of a normal distribution is 3 (Joanes and Gill, 

1998). 

When is the fourth moment of the mean. 

RESULTS 

(26) 

Figure 1 shows the IEEE 30 bus system which is 

separated into three areas. Power must be transferred 

among these areas by three interconnection paths. 

Based on Figure 1, these transaction paths which are 

called T1 (between area 1 and area 2), T2 (between area 

1 and area 3) and T3 (between area 2 and area 3) 

contain several lines. These lines are connected between 

sender buses and receiver buses as shown in Figure 1. 

Power could be transferred between sender and receiver 

areas by these transfer lines. Multilateral probabilistic 

ATC calculation is done for this system and the statistical 

analysis is described for 3 different areas for the IEEE 30 

bus system. 

Statistical analysis was done for ATC based on line 

outage and time varying load. The results are shown in 

Table 1. According to these results, the ATC mean and 

ATC standard deviation comparison calculated by these

Figure 2. Mean comparison for ATC based on line outage and time varying load. 

Figure 3. Standard deviation comparison for ATC based on line outage and time 

varying load. 

two methods for IEEE 30 bus system. Figures 2 and 3 

indicate that the mean of ATC based on time varying load 

is more than the ATC mean based on line outage for all 

transaction paths and the standard deviation for time 

varying load is also less than line outage. In overall, the 

probabilistic ATC based on time varying load is more 

reliable than probabilistic ATC based on line outage. It is 

related to having bigger mean and smaller standard 

deviation. Therefore, it can be concluded that more 

power transferred with more reliability can be contracted 


between seller and buyer of energy when ATC is 

estimated based on time varying load. 

Based on the obtained histogram (Figure 4), it can be 

concluded when line outage occurred, ATC intends to go 

right side of curves (skewness is negative). These 

histograms indicate the ATC do not follow normal curve. 

However in time varying load (Figure 5), the ATC is 

distributed around the mean (skewness is close to zero) 

and their histograms show the ATC follows1 the normal 

curve.


DISCUSSION 

(a) 

(b) 

(c) 

Figure 4. ATC Histogram for Line Outage; (a) Area1 – Area2 (T1), (b) Area1 - 

Area3 (T2), (c) Area2 - Area3 (T3). 

For IEEE 30 bus system, ATC calculations are compared 

against Optimal Power Flow (OPF) (Xiong and Guoyu, 

2005), Benders (Shaaban et al., 2003; Weixing and 

Xiaoming, 2008) and Hybrid Continuous Ant Colony 

Optimization (HCACO) (Guoqing et al., 2008). These are 

optimization methods which considered most of the 

limitations. The percent difference (Diff %) are 

determined for Krylov Algebraic Method (KAM) result 

against each of these methods. 

(27) 

The percentage differences are shown in columns 3, 5 

and 7 of Table 2. Comparing KAM to Benders and OPF,

(a) 

(b) 

Figure 5. ATC Histogram for Time Varying Load; (a) Area1 – Area2 (T1), (b) Area1 - 

Area3 (T2), (c) Area2 - Area3 (T3). 

the amount of ATC has improved except for T2 by 

Benders. Comparing with HCACO, KAM results are less 

than HCACO with 7.78, 0.47 and 6.12 percent 

differences for T1, T2 and T3. The small percent 

(c) 


difference proves that the deterministic ATC results of 

KAM compared well to these methods. Since the amount 

of ATC for T3 calculated by Benders is very small the 

difference between KAM and Benders in this case is big.


Table 2. Deterministic ATC comparison results for IEEE 30 Bus system. 

Transaction Paths 

Benders Diff% OPF 

ATC (MW) 

Diff% HCACO Diff% KAM 

T1 104.19 2.48 101.5 3.45 115.46 7.78 106.81 

T2 103.31 0.35 96.96 4.04 103.43 0.47 102.95 

T3 32.21 39.43 47.59 0.61 45.18 6.12 48.03 

Table 3. ATC Statistical comparison for IEEE 30 bus systems. 

Sending area to 

receiving area 

GA HCACO KAM (Proposed method) 

Mean Standard 

deviation 

Mean Standard 

deviation 

Mean Standard 

deviation 

T1 106.54 2.36 115.46 2.42 106.81 2.61 

T2 98.85 3.50 103.43 2.33 102.93 2.41 

T3 42.82 1.49 45.18 0.45 48.03 0.58 

However, this result is far from the OPF and HCACO with 

38.55 and 33.52% differences. 

The mean and standard deviation of ATC based on 

KAM (Proposed method), GA (Genetic Algorithm) and 

HCACO are shown in Table 3 for IEEE 30 bus system. 

Columns 2, 4 and 6 of this table show the ATC mean of 

T1, T2 and T3 transactions. The related standard 

deviations are also shown in columns 3, 5 and 7 of this 

table. Based on this Table, the mean of ATC for KAM is 

bigger than GA. However its standard deviation is less 

than GA except for T1. In overall, the result of KAM is 

better and more reliable than GA. Compared to HCACO, 

the mean and standard deviations of KAM are smaller 

and bigger. 

Conclusion 

ATC calculation can be divided into two categories, 

deterministic ATC and probabilistic ATC. In this paper 

ATC was calculated for IEEE 30 bus systems which took 

into account the full AC load flow, linear optimization, 

thermal and voltage constraints. The results of ATC 

computation obtained from MATLAB programming were 

used to estimate the ATC by using Minitab software by 

considering time varying load and line outage for power 

system planning. 

To verify the proposed method, the results of 

deterministic and probabilistic ATC for IEEE 30 bus 

system were compared to other methods. The 

deterministic result with small percent difference is close 

to HCACO. However the amount of deterministic ATC for 

KAM is bigger than Benders and OPF. Lower standard 

deviation and higher mean of ATC based on time varying 

load indicate that the time varying load is more suitable 

than line outage to identify the probabilistic ATC. The 

main statistical results for IEEE 30 bus system were also 

compared with GA and HCACO methods. According to 

the results, the ATC standard deviation of KAM is less 

than the GA algorithm and this result is close to HCACO 

which was determined based on random search 

technique. 

REFERENCES 

Adam S (1996). Fundamental Concepts of a Krylov Subspace Power 

Flow Methodology. IEEE Trans. Power Syst., 11(3): 1528-1537. 

Anselmo B, Rodrigues G, Silva MD (2007). Probabilistic Assessment of 

Available Transfer Capability Based on Monte Carlo Method with 

Sequential Simulation. IEEE Trans. Power Syst., 22(1): 484-492. 

Borkowska B (1974). Probabilistic Load Flow. IEEE Trans. Power Appl. 

Syst., PAS93(4): 752-759. 

Christopher C, Michael A (1982). LSQR: An Algorithm for Sparse Linear 

Equations and Sparse Least Squares. ACM Trans. Math. Softw., 

8(1): 43-71. 

Daniel S, Ralph BD (1996). Practical Engineering Statistics. Canada: 

John Wiley and Sons. 

Dopazo J, Klitin O, Sasson A (1975). Stochastic load flows. IEEE Trans. 

Power Syst., PAS-94(2): 299- 309. 

Feller W (1971). An Introduction to Probability Theory and Its 

Applications, Second Edition. New York: John Wiley and Sons. 

Gao Zhou, YM, Li G (2006). Sequential Monte Carlo Simulation Based 

Available Transfer Capability Calculation. Int. Conf. Power Syst. 

Technol., pp. 1-6. 

Gnanadass R, Ajjarapu V (2008). Assessment of Dynamic Available 

Transfer Capability using FDR PSO Algorithm. Elektrika J. Elect. 

Eng., 10(1): 20-25. 

Golub G, Kahan W (1965). Calculating the Singular Values and 

Pseudoinverse of A Matrix. SIAM J. Numer. Anal., 2: 205-224. 

Guoqing L, Zhiyuan L, Hao S (2008). Study of Available Transfer 

Capability Based on Hybrid Continuous Ant Colony Optimization . 

IEEE Conf. Elect. Utility Deregul. Restruct. Power Technol., pp. 984- 

989. 

Huang GM, Yan P (2002). Composite System Adequacy Evaluation 

Using Sequential Monte Carlo Simulation for Deregulated Power 

Systems.IEEE Power Engineering Society Summer Meeting. 

Ioannis K A (2007). Computational Theory of Iterative Methods (Vol. 

15).Elsevier.

Joanes DN, Gill CA (1998). Comparing Measures of Sample Skewness 

and Kurtosis. J. Royal Stat. Soc. (Series D): The Stat., 47(1): 183- 

189. 

Jonathan WS (2007). Stochastic-Algebraic Calculation of Available 

Transfer Capability. IEEE Trans. Power Syst., 22(2): 616-623. 

Sauer P (1977). A Generalized Stochastic Algorithm.W.Lafayette: Ph.D 

Thesis, Purdue University. 

Shaaban M, Li WX, Yan H, Ni YX, Wu FF (2003). ATC Calculation with 

Steady-State Security Constraints using Benders Decomposition. 

IEEE Proc. Generat. Transm. Distr., pp. 611-615. 

Stahlhut JW, Heydt GT, Sheblé GB (2005). A Stochastic Evaluation of 

Available Transfer Capability. IEEE PES General Meeting. 

Weixing AL, Xiaoming M (2008). A Comprehensive Approach for 

Transfer Capability Calculation Using Benders Decomposition. 

DPRT, pp. 301-306. 


Xiong P, Guoyu (2005). Available Transfer Capability Calculation 

Considering Voltage Stability Margin. Elect. Power Syst. Res., 76(1- 

3): 52-57. 

Yajing G, Ming Z, Gengyin L, Yuekui H, Limin X, Rui L (2005). Monte 

Carlo Simulation Based Available Transmission Capability 

Calculation. IEEE/PES Transm. Distr. Conf. Exhibit. Asia Pac., pp.1- 

6.



DOI: 10.5897/SRE12.097 



Seasonality of non-timber forest products in the Kupe 

mountain region of South Cameroon 

Ngane B. K.*, Ngane E. B., Sumbele S. A., Njukeng J. N., Ngone M. A. and Ehabe E. E. 

Institute of Agricultural Research for Development (IRAD), Ekona Regional Research Centre, Buea, South West Region, 

Cameroon. 

Accepted 23 April, 2012 

Non-timber forest products (NTFPs) are economically-important commodities, especially in the humid 

and semi-humid regions of many sub-Saharan countries. However, in many of these countries, there is 

little information on the seasonality characterising their availability and collection, rendering difficult 

the development of meaningful packages for their sustainable use and valorisation. A study was 

therefore conducted in randomly selected villages in three zones around the sub-montane forests of 

Mount Kupe in South West Cameroon, in order to evaluate the effects of seasons on the collection and 

availability of these products. Data were collected through administration of semi-structured 

questionnaires and a rapid rural appraisal. Most NTFPs (85%) showed seasonality to different extents, 

in availability and abundance. Although, considerable differences were observed in the proportion of 

respondents who collected NTFPs during the dry and rainy seasons, no such difference could be 

attributed to the climatic or geographical zones. Half of the respondents collected NTFPs during the 

rainy season when there was relatively less work in their farms. About one-sixth (15.2%) collected 

NTFPs during the dry season (peak agricultural period) and a third (35%) collected some NTFPs all year 

round. 

Key words: Cameroon, humid forest, non-timber, seasonality, collection. 

Introduction 

Until recently, timber products like industrial and derived 

sawn timber, wood chips, wood-based panel and pulp 

have been considered as the major or even the most 

important forest products due to their income-generating 

capacity. Meanwhile, other forest products, especially 

those used as food, medicinal plants, household and 

agricultural implements, etc. have traditionally been considered 

of minor or secondary importance (Makon et al., 

2005). The importance of these later materials, extracted 

from forest ecosystems, has increased considerable 

today following their use within households and as they 

are equally traded for their social, cultural or religious 

significance (Ndoye and Tieguhong, 2004; Ngwasiri et 

al., 2005). Unlike some authors who term these materials 

*Corresponding author. E-mail: nganekoben@yahoo.co.uk. Tel: 

(+237)75936380. 

as non-wood forest products (FAO, 1991; Sunderland 

and Oboma, 1999; Vantomme, 1999), the term nontimber 

forest products (NTFPs) used by other authors 

(Falconer, 1992; Kio and Abu, 1993; Okafor, 1989) is 

preferred in this study since some of the products such 

as axe handles, fuel wood and chewing sticks contain 

lignin (wood). 

Most NTFPs are not domesticated and are just 

collected from the wild. They are often not available yearround 

or when available, are in limited quantity. The 

exploitation of NTFPs could therefore be regarded as a 

seasonal activity that contributes to improving the 

livelihoods of rural populations where prices of major 

agricultural products are low, fluctuating and seasonal. 

Agriculture is the mainstay of the people of the Kupe 

area. Farming is carried out mainly for subsistence and 

the traditionall farming system is shiefting cultivation 

which is wasteful to land. Arable crops, tree crops and 

animals are being farmed. Major arable crops are

Table 1. Market value of some important NTFPs in Southern Cameroon. 

Species name 

% of harvest 

sold 

Total revenue 

(CFA) in 4 months 

Total revenue ($ US) 

in 4 months 

Coula edulis 5 15000 3.1 

Cola acuminate/nitida 30 2500 5.1 

Cola pedidota 30 3800 7.8 

Dacryodes edulis 10 11000 22.5 

Dacryodes macrophylla 10 2000 4.1 

Elaieis guinensis ‹1 2800 5.7 

Garcinia lucida 10 400 0.8 

Irvingia 

wombolu 

gabonensis/ 

20 79000 161.6 

Ricinodendron heudelotti 100 5800 11.9 

Strophanthus gratis 100 32500 66.5 

cocoyam, plantains, banana, cassava, yam, maize, 

cabbage, pepper and okro. Tree crop agriculture involves 

the growing of cacao, coffee and oil palm. Animal farming 

includes the rearing of goats, sheep and birds. Fishing of 

tadpoles and frogs is carried out in in-land water bodies. 

Disturbing and persistent drop in the prices of cocoa 

and coffee (major export crops of the area) for two 

decades now have further worsened poverty levels in this 

area. However the sale of food crops now constitutes a 

major income source particularly as foreign buyers from 

mostly Gabon and Equitorial Guinea have flooded the 

market. NTFPs usually collected in the study area are 

used for food, medicine, household utensils and 

construction material. 

Due to slightly different vegetation and climatic 

conditions in the three zones considered in the study, it is 

expected that species variation and variation in time of 

production of NTFPs will be observed (Leakey et al., 

2000). For example, off season Dacryodes edulis could 

be found in Kupe East. Also Gnetum spp. (Figure 4L) 

which is of local, national, regional and international 

significance could be found only in Kupe East with higher 

temperatures throughout the year. With local variation in 

the vegetation and climatic conditions in the study area, it 

is possible that a particular product could be obtained 

from different species in different zones. 

Although most of the NTFPs are collected for 

household consumption, some are traded to suppliment 

the cash income of households. While NTFPS such as 

Ricinodendron heudelotti and D. edulis are traded locally 

in village markets, others such as Irvingia spp. and 

Gnetum spp. are taken to city markets and are bought by 

buyers who export to Nigeria, Europe and the United 

States of America. 

In the Kupe area, collection of NTFPs is an income 

diversification activity which is risk-reducing and at the 

same times a way of coping with constraints in the main 

income generating activity of the area, agriculture. 

Important NTFPs such as Garcinia cola (Figure 4a), Piper 

Ngane et al. 1787 

guinense (Figure 4E, F and I) and R. heudelotti that play 

a major role in generating income for the local people are 

not available throughout the year. The scarcity of such 

NTFPs during off-seasons leads to food insecurity and 

economic vulnerability. This adversely affects the socioeconomic 

life of the community as malnutrition and 

associated diseases coupled with poverty reduce the 

living standard of the people. 

The absence of a particular medicinal NTFP such as 

Afromonium melegueta (Figure 4J) at a certain period of 

the year can also result to serious health problems in the 

communities that depend on it for the cure of various 

ailments. Dijks Van (1999) showed that NTFPs collection 

and production can provide a source of livelihood as 

shown on Table 1. 

In such an area where NTFPs play a major role in 

enhancing status and wellbeing of the local people, 

where there is ethnic diversity, local variation in climatic 

conditions with consequent variation in use and 

availability of NTFPs during different seasons of the year, 

there is need for relevant, reliable and timely information 

on their seasonal availability and estimates of their stocks 

if adequate forecasts of use, marketing and sustainability 

are to be made. 

The objective of this study was therefore to provide 

information on the true value and significance of nontimber 

forest products during different seasons of the 

year in the Mount Kupe area. More specifically this study 

attempts to: 

i. Identify the various types of non-timber forest products 

in the study area during the different seasons. 

ii. Assess the factors affecting the involvement of local 

populations in their collection. 

iii. Proffer recommendations on management strategies 

that will mitigate the effect of seasonality of these 

products. 

Based on the objectives of this study the hypothesis


tested were as follows: 

i. That the availability of NTFPs depends on different 

seasons of the year. 

ii. That the availability of NTFPs depends on zones. 

iii. That NTFPs contribute to the socio-economy of the 

local people during different seasons of the year. 


Survey locus 

The Kupe area straddles two regions of Cameroon, the South West 

and Littoral regions. The population that inhabits the Kupe area 

represents an array of socio-cultural, economic and political 

identities possessing a diverse ethnic originality made up of various 

indigenous Bantu clans (Muetuck, Mbo, Manehas, Muasundem 

etc), settlers from the North West Region and immigrant Bamilikis 

from the West Region of Cameroon. This people make up a 

population of about 100,000. 

Forests that make up the Mount Kupe are mostly sub-montane 

equatorial rain forest and are known to support a wide variety and 

abundance of resources which are of huge importance to the 

people that inhabit the area for livelihood improvement, recreation, 

cultural promotion, etc. Forests resources present include timber, 

NTFPs, ethno-botanical resources, etc. The area is also known to 

have huge eco-tourism potentials as well as very rich cultural 

values. Moreover the quality of the forest provides for undeniable 

environmental services including water catchments. 

The survey was conducted through out the year 2008 in the 

Mount Kupe area located within the humid forests of Southern 

Cameroon. The Kupe Mountain spans the borders of the South 

West and Littoral Regions of Cameroon. The area, home to the 

Bakossi tribe and a high proportion of non-indigenes, could be 

partitioned into three zones populated by three different clans with 

distinctively different eating habits: Manehas in the East 

(Francophone), Muetuck in the West (Anglophone) and 

Muasundem in the North- Anglophone (Ejedepang-Koge, 1986). 

These zones are different with respect to their altitudes, 

geographical positions, climate and vegetation. 

The difference in altitudes and geographical position has led to 

differences in the climate as well as the vegetation of the different 

zones. The climate of this area is of the equatorial or transitional 

tropical type. Despite some local variations, there is a zonal 

uniformity of high and constant temperatures. Rainfall varies from 

2500 mm per annum in Kupe East, to about 5000 mm per annum in 

the entire area. Minimum monthly temperatures vary from 19 to 

21°C; though much lower in Kupe West (2064 m) and maximum 

from 30.55 to 35.20°C. Relative humidity is about 85%. 

Generally two main seasons could be distinguished separated by 

an intermediate season. (i) Mid-June to October: a major rainy 

season of overcast and misty weather, continuous rains and 

drizzling; (ii) Mid-November to Mid March: a major dry season of dry 

foggy weather, harmattan and diffuse radiance, (iii) Mid October to 

mid-November/mid-March to mid June: intermediate/transition 

season of changeable weather, storms and rain showers alternating 

with bright intervals. 

Sampling of respondents and data collection 

The studied area was stratified based on the zones around the 

Kupe mountain (East, West and North of the mountain). Three 

clans were randomly selected from the entire Kupe area, and three 

villages randomly selected from each clan, as shown as follows: 

i. Kupe East zone (Manehas clan): Nlohe, Lala and Kola villages 

(French speaking villages); 

ii. Kupe West zone (Muetuck Clan): Nyasoso, Ngusi and Atop 

villages (English speaking villages); 

iii. Kupe West zone (Muasumdem clan): Mpako, Tape-Etube and 

Bendume villages (English speaking villages). 

A semi structured questionnaire was used as the main survey 

instrument for information collection. It was designed in five parts A 

to E. Section A was designed to elicit information on the 

background of the villages. Section B dealt with the biodata of the 

respondents. Section C sought to elicit information on the 

identification of NTFPs. Section D was to gather information on 

seasonal variation in the collection of NTFPs and section E dueled 

on the socioeconomic contribution of NTFPs. 

Before the actual administration of the questionnaire a 

reconnaisance survey was undertaken during which some 

background information on the study area was collected. A pre-test 

survey was also undertaken. During this exercise test questionnaire 

were randomly administered in three villages which though different 

from the villages selected for the actual study were within the same 

locality. The completed test questionnaire were collated and 

analysed. Adjustments were made before the final production of 

questionnaire before the survey proper was undertaken. 

Ten percent of the estimated number of households in the 

selected villages was calculated (Hannagan, 1989). 280 

questionnaire were filled and 250 returned as follows: 66, 87, and 

97 for Kupe East, Kupe North and Kupe West respectively. In 

addition to this, a rapid rural appraisal (RRA) was carried out - using 

transect walks, structured direct observations and key informant 

interviews. 

Data analyses 

The data were statistically analysed for the sources of variance on 

a randomised complete block design using the M-stat software 

(Alika, 1997). This was complimented by descriptive statistics using 

the Sphinx package. 

Results 

This survey showed that non-timber forest products were 

obtained from about sixty plant species and twenty 

animal genera (Tables 2 to 4). The majority of 

respondents collected NTFPs during the rainy season 

(Figure 2). On the average for the three zones, about half 

(50.4%) of the respondents collected NTFPs during the 

rainy season, while 15.2% collected NTFPs during the 

dry season. Also, 34.4% collected some NTFPs all the 

year round. Significant differences (P < 0.05) were 

observed between the number of respondents collecting 

NTFPs during the rainy and dry seasons but there was 

no significant difference between those collecting in the 

rainy season and those collecting all year round. There 

was a significant difference in the number of respondents 

collecting NTFPs during the different seasons in the 

different zones of the study area. The main reason for 

much collection of NTFPs during the rainy season was as 

a result of less farm work during this period (Table 5); 

though most of the NTFPs show seasonality in their 

collection (Tables 2, 3 and 4). Some human food NTFPs

show stark variation in collection during the different 

seasons of the year (Table 2). For example, though 

bushmeat (game) is generally harvested more frequently 

during the rainy season, pestilential species were caught 

relatively more frequently during the dry season. In the 

Eastern side of Kupe, fruit production was much later 

(end of season) and off-season fruit were available for 

collection. Construction materials show much less 

seasonality. However most construction and maintenance 

of houses, fences and huts takes place during 

the dry season (Table 3). Non-timber forest products also 

provide a source of livelihoods for rural people. This 

employment is basically on part-time basis (Figure 3). 

The situation was similar in the different zones but more 

full-time employment was provided by NTFP activities 

during the rainy season than during the dry season. 

Discussion 

The reason for collection of NTFPs in the different 

seasons was tested and revealed that most respondents 

collect NTFPs during the rainy season because of less 

work in their farms. This was also observed by 

Sunderland and Oboma (1999). 

This can further be attributed to the fact that during the 

dry season most labour is devoted to agricultural 

activities, while during agricultural slack periods much 

time is allocated to NTFP collection. This sometimes 

leads to over harvesting and unsustainable harvesting of 

NTFPs. Biliso and Lojoly (2006) also posited that there 

are no other significant sources of income during the 

rainy season. During the peak of the rainy season 

(hungry period), stored food supplies dwindle, most food 

crops on the farm are yet to mature and as such there is 

scarcity of food. NTFPs are important during this period 

as substitutes for staple foods, accompaniments and 

snacks to supplement the food crop supplies. When not 

available the local people are deprived of important food 

nutrients, medicines and household income. This has a 

very serious adverse effect in Kupe West and Kupe North 

than in Kupe East where off-season products are 

available and commercialisation of NTFPs more lucrative 

due to access to major city markets. 

Some respondents showed that harvesting periods of 

most NTFPs in Kupe coincide with school resumption 

(September in rainy season), hence NTFPs are highly 

needed to raise income for fees, uniforms and purchase 

of books. Buyungu (2000) also observed that this was 

also the situation in other West African countries like 

Ghana, Nigeria and Cote d’Ivoire. Some NTFPs such as 

D. edulis and forest vegetables are very much perishable 

hence more seasonal than bush mango (Irvingia 

gabonensis and wombulu Figure 4D) which can be dried 

to be used when not in season. This has also been 

observed by Leakey et al. (2000). It was also observed 

that Irvingia gabonensis was available in the rainy season 


(June-September) while Irvingoa wombulu was available 

during the dry season (October to February) making the 

product, ogbono available all year round. It was also 

observed that off-season fruits could be found in the 

Eastern side of Mount Kupe probably due to dryer 

climatic conditions in the leeward side of the mountain. 

Another important reason for much NTFPs collection 

during the rainy season is that very important NTFPs 

such as D. edulis (plum or safou), G. cola (bitter kola, 

Figure 4G) and Cola acuminate and nitida (kolanuts, 

Figure 4A) mature during the rainy season. 

The study showed that in the heart of the dry season, 

the availability of snails dwindles as they go into 

aestivation and are sometimes destroyed by wild fires. 

More forest vegetables were collected during the dry 

season as a result of scarcity of vegetables from farms. 

According to Hart et al. (2005), the harvesting of forest 

vegetables is more during the dry season because these 

serve as alternatives to cultivated vegetables which are 

perishable and scarce at this period. Though more game 

was obtained during the rainy season, pestilential species 

of bushmeat are caught during the dry season when their 

habitats are destroyed during land preparation for the 

next planting season. In this study, it was observed that 

the availability of some forest spices such as 

Ricinodendron heluidothii (njangsang) and Tetrapleura 

tetraptera (four corner or essisang) show very little 

seasonal variation in availability because they generally 

keep well, if dried. Value is added when preserved and 

are sometimes exported to neighbouring Nigeria. This 

contributes immensely to improving the living standard of 

the local people. 

Palm wine consumption largely depended on seasonal 

availability and perishability and is mostly drunk during 

the dry season when tapping is easy, without dilution 

from rain water. Local gin (locally called “afofo” and 

“odontol”) mostly produced from palm wine is drunk more 

during the rainy season when palm wine is relatively 

scarce. People tend to drink more of this local gin during 

the rainy season to have themselves warmed up. 

The collection of non-timber forest products used as 

medicine, construction materials and household items 

was much less seasonal because the parts mostly used 

for these purposes are the bark (cortex) and wood which 

are less perishable and are available throughout the year. 

Also in the case of some species like bitter cola (G. cola) 

and ‘ngou’ (Garcinia lucida) whose fruits are used as 

medicine, the bark is taken as a good substitute for 

treating the same ailments when the fruits are not in 

season. This also has a destructive effect on the tree 

which is sometimes girdled to death, especially as this 

occurs even in the dry season. Much harvesting of 

medicinal plants was observed in Kupe North because 

Croton spp. (“Ndume”, Figure 4B) which is considered by 

the Bakossi ethnic group as the most important medicinal 

plant is abundant in Bendume village from which the 

village derived her name (Bendume meaning plenty of


Table 2. Seasonality of major food NTFPs in the Kupe area. 

Source Product 

Usefulness in the study area 

Rai 

ny 

Kupe West Kupe North Kupe East 

Dry 

All 

year 

Rainy Dry All year Rainy Dry All year 

(a) – Food of plant 

origin 

Elaeis guineensis 

Palm oil and kernel Cooking oil 2 3 2 1 2 1 2 3 2 

Palm wine Alcoholic drink 1 3 1 1 3 1 1 1 1 

Raphia hookery Palm wine Alcoholic drink 2 3 2 1 2 1 3 1 1 

Spices and condiments 

Ricinodendron heudelotii Njasang Condiment, vitamin c 3 1 1 3 1 1 2 1 1 

Irvingia gabonensis Ogbono condiment 2 0 0 2 0 0 1 0 0 

Irvingia wombulu Ogbono condiment 0 2 1 0 1 0 0 1 0 

Piper guinense Bush pepper Spice 2 3 2 0 1 0 1 1 1 

Tetrapleura tetraptera Essisang spice 2 3 1 1 2 1 1 1 1 

Fruits/seeds 

Dacryodes edulis Plum 

Snacks and supplement for meat 

(protein) 

3 1 1 2 0 0 3 1 1 

Cola acuminate 5 piece kola Stimulant and aphrodisiac 3 1 0 1 0 0 1 0 0 

Cola nitida 1 piece kola Stimulant and aphrodisiac 0 0 0 0 0 0 2 1 1 

Garcinia cola Bitter cola 

Stimulant and aphrodisiac, snake 

repellent, treatment of coughs and 

hepatitis 

3 1 0 1 0 0 0 0 0 

Aforsthyrix sp. Bush onion Spice 2 1 1 2 0 0 1 0 0 

Mushrooms and 

vegetables 

Agaricus campestries Kokobiako Source of protein and condiment 3 1 1 2 1 0 2 1 0 

Gnetum africanum Eru Vegetable 0 0 0 0 0 0 1 2 2 

Gnetum buchulzianum Eru Vegetable 0 0 0 0 0 0 1 2 2 

Other forest vegetables 

(b) – Food of animal 

origin 

Insects 

2 3 2 2 2 1 1 2 1 

Apis malifera Honey bee Source of protein 1 3 1 1 3 1 1 2 1 

Rhynchophorus 

phoenicis 

Maggot Source of protein 1 2 1 2 1 2 1 3 1 

Bush meat (game) 

Cricetomys gambianus Bush rat Source of protein 3 2 1 3 2 2 3 2 1

Table 2. Cond. 


Thryonomys 

swinderianus 

Cane rat Source of protein 2 3 2 2 3 2 2 2 1 

Cephalopus spp. Duikers Source of protein 3 2 2 2 1 1 2 1 1 

Atherurus africanus Porcupine Source of protein 3 2 2 2 3 2 2 3 2 

Monkeys Source of protein 3 2 2 3 1 1 3 1 1 

Potomochoerus porcus Bush pig Source of protein 2 1 1 2 1 1 1 1 1 

Fishes Source of protein 0 0 0 1 1 1 2 1 1 

Birds Source of protein 1 1 1 1 1 1 1 1 1 

Snakes . Source of protein 3 2 2 3 2 2 2 1 1 

Snails 

Archachatina. Archatina Snail Source of protein 3 1 1 1 1 1 1 0 0 

Archachatina marginata. Snail Source of protein 0 0 0 0 0 0 0 .0 0 

0 = none; 1 = very few; 2 = few; 3 = many. 

Table 3. Seasonality of major non-food NTFPs in the Kupe area. 

NTFP and source Usefulness in the study area 

Kupe West 

Rainy Dry All year 

Kupe North 

Rainy Dry All year Rainy 

Kupe East 

Dry All year 

Cordia sp. Construction of houses 3 3 1 3 2 2 2 3 2 

Elaeis. Guineensis (palm fronds) Construction of houses 1 3 2 1 3 1 1 3 1 

Raphia spp. (palm fronds) Construction of houses 1 2 1 1 2 1 1 2 1 

Thaumatococcus spp. (ngongo leaf) Wrapping leaves 3 2 2 2 2 2 3 3 2 

Neubouldia laevis Live fencing 3 3 3 2 2 2 1 1 1 

Household implements 

Bambusa sp. (Indian bamboos) Construction of houses 2 2 2 3 3 3 2 2 2 

Cordia sp. Mortar 3 3 1 3 2 2 2 3 2 

Elaeis guineensis (palm oil) brooms 1 1 1 1 1 1 2 3 2 

Calamus spp. (cane) Can chairs baskets, beds 3 3 3 2 2 2 2 2 2 

Acacia spp. (gums) gum 0 0 0 0 0 0 1 1 1 

Irvingia spp. carving 1 1 1 1 0 0 0 0 0 

Ptericarpus sauyouxii (camwood) Mortar, cosmetics 2 2 2 2 2 2 2 2 2 

Firewood fuel 1 3 2 1 2 2 1 2 1 

Albizia spp., Irvingia sp. etc. 

0 = none; 1 = very few; 2 = few; 3 = many.


Table 4. Seasonality of major medicinal NTFPs in the Kupe area. 

NTFP and source 

Medicine 

Croton spp. 

Elaeis guineensis 

(kernels) 

Gnetum africanum 

Usefulness in the 

study area 

Most are marketed: 

Chase evil spirits, 

stomach ache, well 

being of pregnant 

women 

Babies welfare and 

cosmetics 

Cure for spleen, bed 

wetting, stomach 

ache, drunkenness 

Rain 

y 

Kupe West Kupe North Kupe East 

Dr 

y 

All 

year 

Rainy Dry 

All 

year 

Rain 

y 

Dry All year 

1 1 1 3 3 3 0 0 0 

1 1 1 1 1 1 3 3 3 

0 0 0 0 0 0 1 2 2 

Gnetum buchulzianum 0 0 0 0 0 0 1 2 2 

Enanthia chlorantia 

(Chloroquine plant) 

Azadirachta indica 

(Neem) 

Garcinia cola (bitter 

kola) 

Allanblankia floribundia 

Prevention and cure 

for malaria, added to 

local gin 

Treatment of malaria, 

preservation of 

stored crops 

Cough, hepatitis, 

aphrodisiac 

Treatment of 

toothaches and 

diarrhoea 

2 1 1 3 3 3 0 0 0 

1 1 1 1 1 1 3 2 2 

3 1 0 2 0 0 3 1 1 

3 1 1 0 0 0 3 2 1 

Albizia sp. (essang) Malaria 2 2 2 2 2 2 1 1 1 

Garcinia lucida 

(meckak) 

Tooth ache 3 3 3 2 2 2 2 2 2 

Garcinia mannii (ngou) Tooth bush 3 3 3 2 2 2 2 2 2 

Maesopsis eminii 

(Rhamnacea) 

Afromonum melegueta 

(alligator peper) 

Cosmetics 

Pterocarpus soyauxii 

(camwood) 

Gonorrhoea and 

other bacterial 

infections 

2 2 2 1 1 1 0 0 0 

Charms, aphrodisiac 3 3 3 3 3 2 2 2 2 

Rubbed on women 

after delivery 

3 3 3 2 2 2 2 2 2 

Tectona grandis (teak) Bathing sponge 1 1 1 1 1 1 0 0 0 

Chewing stick 

Garcinia mannii Tooth brush 3 3 3 2 2 2 0 0 0 

Massularia acuminata Tooth brush 3 3 3 2 2 2 1 1 1 

Animals 

Chamelion 

(Chamaeleon spp.) 

Reduce shock from 

fright after accidents 

3 3 3 2 2 3 1 1 1

Table 4. Cond. 

Pocupine (Atherurus 

africanus) 

Bush fowl (Francolinus 

cameroonensis) 

Giant rat (Cricetomys 

cambianus) 

Grass cutter hair 

(Thryonomys 

swinderianus) 

Easy child birth in 

women 

0 = none; 1 = very few; 2 = few; 3 = many. 


2 2 2 2 2 2 1 1 1 

Prevent child birth 1 1 1 1 1 1 1 1 

Protection against 

witchcraft, poisons 

and charm to win the 

love of women 

3 3 3 1 3 2 1 1 1 

Healing of eyes 2 2 2 2 2 2 0 0 0 

� Table 5. Reasons for seasonality in collection of NTFPs. 

� 

Reason Kupe East % Kupe North % Kupe west % Total % X 

Little farm work 3.3 28.4 38 32.7 45 38.8 11.6 46.4 68.7 

Availability 2.0 20.8 34 35.4 4.2 43.7 9.6 38.4 32 

Weather 13 34.2 15 39.5 10 26.3 38 15.2 12.7 

Total 66 26.4 87 34.8 97 38.8 250 100 

Ndume). 

Much firewood and yam stakes are collected during the 

dry season since land preparation for the next farming 

season involves clearing of bushes. Some respondents 

revealed that a little more construction material is used 

during the dry season. At this time houses and other huts 

are constructed using poles, palm frond petioles or 

bamboo stems and twines. During the rainy season, 

some of these constructions require maintenance 

andconstruction materials are again needed to fortify 

them. 

The study revealed that a particular product could be 

obtained from different species in different zones. For 

instance, palm wine is obtained from both Elaeis 

guineensis and Raphia hookery in all the clans, but the 

latter is more commonly found in fresh water swamps in 

Kupe West and more available in the rainy season as it is 

easier to tap (less readily diluted by rain water during 

tapping). 

Though collection of NTFPs provides employment to 

the inhabitants of the area, agriculture is the mainstay of 

the people of Kupe. The people are more involved in 

farming activities during the dry season than during the 

rainy season. Little time is spent on gathering of NTFPs 

during the period of crop cultivation. However, during 

agricultural slack periods the collection, processing and 

marketing of NTFPs becomes the major occupation of 

some rural dwellers. More fulltime employment was 

observed in Kupe West and Kupe North than in Kupe 

East (Figure 3). This is probably because of the 

abundance of the products in Kupe West and Kupe North 

than in Kupe East (Tables 2, 3, and 4). 

CONCLUSION 

This work has thrown some light on the availability and 

collection of non-timber forest products in Kupe. These 

products show seasonality in availability and collection 

with most of them being collected during the rainy season 

when there is little work in the farms and when food crops 

are yet to mature. The duration of the rainy season 

(7months) in the study area is more than that of the dry 

season (5 months). This is another reason why more 

NTFP collection takes place during the raining season. 

Collection of NTFPs is an income diversification activity in 

The area and is also a risk-reducing strategy against 

household income fluctuations. 

It was observed that during particular periods of the 

year a particular NTFP will exist in abundance and 

sometimes in excess while there is acute scarcity during 

another period of the year. This seasonal nature of 

NTFPs has been an enormous setback in the utilisation 

of these important resources. To enable the local people 

and other actors benefit from the exploitation of NTFPs, 

research on production of off-season varieties should be


Atop 

Ngusi 

Kupe 

West 

… Mount Kupe farm/forest boundary 

… Clan boundary 

� Study villages and forest 

Mpako 

Tape- 

Etube 

Bendume 

Mount 

Kupe 

Figure 1. Location map of the study site (Mt. Kupe area). 

Figure 1. Location map of the study site (Mt. Kupe area). 

encouraged. Selection for all-year-round production for 

important NTFPs such as D. edulis and Irvingia spp. can 

certainly have considerable impact on the availability of 

Nyasoso 

Nlohe 

Kupe 

East 

Kola 

Lala 

products as there is typically some within-species 

variation in phenology with some trees flowering and 

fruiting outside the main season. Artificial and deliberate

Percentage of respondents 

60 

50 

40 

30 

20 

10 

0 

Seasonality of NTFP Collection in the Kupe Area 

Kupe East Kupe North Kupe West 

Zone 

� 

� Figure 2. Seasonal variation of NTFP collection in the study area. 

Percentage of respondents 

100 

90 

80 

70 

60 

50 

40 

30 

20 

10 

0 

Employment pattern of NTFPs 

Part-time Full-time Part-time Full-time 

Rainy season Dry season 

Pattern and seasons 

Rainy 

Dry 

All year 

K West 

K North 

K East 

Figure 3. Effect of geographical zone and different seasons of the year on employment patterns in 

the survey zone. 

actions such as improved preservation techniques, 

irrigation and incorporation of NTFPs in agroforestry 

systems and organic gardens can be important in 


providing off-season products. This will be very useful 

today in the mitigation of the effect of climate change on 

availability, collection and management of NTFPs.


A B C 

D E 

G H 

J K L 

Figure 4. Some common NTFPs in the Mount Kupe area found in the Tombel Town Market place. 

REFERENCES 

Alika JE (1997). Statistics and Research Methods. 1 st Edition, Ambik 

Press Publishers, Benin City, Nigeria. 

Biliso A, Lejoly J (2006). A Study of the Exploitation and Markets of 

Non-Timber Forest Products in Kinshasa. Tropicult, 24(3): 183-188. 

Buyungu PM (2000). The Contribution of Plums (Dacryodes edulis) in 

the diversification of Agriculture in West Africa. In kengue J, Capseu 

C, Kayem GJ (eds.), Third International Seminar on the Valorisation 

of Plums and other Non-Conventional Oil Crops. Yaounde, 3-5 

Octobre 2000. DFID, CTA IFS CIRAD, Pub. African University Press 

Yaounde, 638 p. Programme (E.D.A.D.P.). Ministry of Agriculture, 

Benin City. Edo State. 

Dijks Van (1999). An Assessment of Non-wood Forest Product 

Resources for Development of Sustainable Commercial Extraction, 

In: Sunderland TCH, Clark LE, Vantomme P (Eds.) Non-wood forest 

F 

I 

products of Central Africa: current research issues and prospects for 

conservation and development. Based on the outcome of the 

International Expert Meeting on Non-wood Forest Products in Central 

Africa, held at the Limbe Botanic Garden, Cameroon, 10-15 May 

1998. FAO, URL: htt://www.fao.org/do crep/x216 le 32. Htm . 

Ejedepang-Koge SN (1986). The Tradition of a People Bakossi: A 

Historico-Socio Anthropological Study of one of Cameroon’s Bantu 

Peoples. Virginia, USA: ARC Publications, 362: 42-45. 

Falconer J (1992). Non-Timber Forest Products in Southern Ghana. 

Wilson et al., (eds). Publ. National Resources Institute, Chatham M. 

K. ME4 4TB, UK., pp. 6-7. 

Food and Agricultural Organisation of the United Nations (FAO) (1991). 

Marketing Non Wood Forest Products in Developing Countries. 

UNASYL. J., 46(1): 37-41. 

Hannagan TJ (1989). Mastering Statistics. 2 nd edition, MacMillan 

Education. Ltd, 272: 43-57.

Hart AD, Azubuike CU, Barimala IS, Achinewhu S C (2005). Vegetable 

consumption pattern of households in selected areas of the Old 

Rivers State in Nigeria. Afri. J. Food, Agric. Nutri. Dev., 5(1): 18-25. 

Kio PRO, Abu JE (1993). Environmental accounting mechanism for 

reconciling land use pressure on forests; In: PRO Kio. (Ed), 

UNEP/CIFOR Project for West/Central Africa, University of Ibadan, 

Ibadan, Commer. For. Rev., 72(4): 272-278. 

Leakey RR, Greenwell BP, Hall MN (2000). Domestication of 

indigenous fruit trees in West and Central Africa: In: kengue J, 

Kapseu C, Kayem GJ (eds.), Third International Seminar on the 

Valorisation of Plums and other Non-Conventional Oil Crops. 

Yaounde, 3-5 Octobre 2000. DFID, CTA IFS CIRAD, Pub. African 

University Press Yaounde, 638: 73-92. 


Ndoye O, Tieguhong JC, (2004). Forest Resources and Rural 

Livelihoods: The Conflict between Timber and Non-timber Forest 

Products in the Congo Basin. Scan. J. Forest. Res., 19: 36-44. 

Ngwasiri C, Djeukam N, Vabi M (2005). Lagislative and Institutional 

Instruments for the Sustainable Management of Non-timber Forest 

Products (NTFPs) in Cameroon. Past, Present and Unresolved 

Issues. Community Forestry Development Project, Yaoundé, 

Cameroon. 

Okafor JC (1989). Agroforestry Aspects. Appendix 2 of the Cross River 

National Park Oban Division Plan for Developing the Park and Its 

Support Zone. Godalming. W.W.F, U.K, pp. 7-90.



DOI: 10.5897/SRE12.223 



Adaptive message size routing strategy for delay 

tolerant network 

Qaisar Ayub, M. Soperi Mohd Zahid*, Sulma Rashid and Abdul Hanan Abdullah 

Department of Computer System and Communication, Faculty of Computer Science and Information System, 

Universiti Teknologi Malaysia (UTM) Skudai - Johor, 81310, Malaysia. 

Accepted 3 May, 2012 

Delay tolerant network (DTN) is a kind of computer network that suffer from the frequent 

disconnections, network partitioned and unstable network connectivity, therefore maintaining an 

uninterrupted route from source to destination is not possible. Therefore, the transmission of message 

is achieved via intermediate nodes by adopting a novel transmission mechanism called store-carry and 

forward where node stores the incoming message in its buffer, carries it while moving and forward it 

when it comes in the transmission range of other nodes. DTN routing protocols can be either single 

copy or multi copy. In single copy protocols, the node forwards the unique copy of message along a 

single path. These protocols suffer the long delivery delay. In multi copy protocols, the node diffuses 

multiple copies of same message along dissimilar paths. Thus, message can reach destination via more 

than one path. However, the replication process consumes high volume of network resources such as 

buffer space, bandwidth and node energy. The probabilistic routing strategies for instance PRoPHET 

Protocol minimizes the consumption of resources and forwards a message to a custodian by using a 

metric of delivery probability. The probability describes the suitability of a node to meet the destination 

of message. However, when node mobility pattern is not symmetric the probabilistic computations 

cannot predict the accurate forwarding decision. In this paper, we have proposed a novel message 

forwarding strategy called Adaptive Message-Size Routing strategy (AMRS) by which a node handovers 

the copy of message to its neighboring nodes by using a metric named as mean threshold (MTH). We 

have compared the performance of AMRS with Epidemic and PRoPHET routing protocols. The 

proposed routing strategy has performed better in terms of maximizing delivery probability while 

minimizes message drops and number of transmissions. 

Key words: Store-carry-forward, routing protocols, delay tolerant network, algorithm. 

INTRODUCTION 

Ad hoc routing protocols for instance table driven and ondemand 

(Johnson et al., 2001; Ahmedy et al., 2011) 

establish an end-to-end path prior to the transmission of 

data. However, these solutions are not applicable in 

applications like wild life monitoring, deep space 

*Corresponding author. E-mail: soperi@utm.my. 

Abbreviations: MTH, Mean threshold at active node; BU, 

buffer used by the active node; Mn, Number of messages at 

active node; MDest, message destination; Msize, message size; 

APN, Appetizer Node; ACN, active node. 

communication, military and sensor network where the 

node mobility causes dynamic topology changes and 

frequent disconnections. As a result, source and 

destination cannot sustain the uninterrupted path. 

The delay tolerant network (Fall and Farrell, 2008) 

provides the transmission of data for such scenarios by 

applying the novel message diffusion strategy referred to 

as store, carry and forward. According to this approach, 

the node stores the incoming message in its buffer, 

carries it while moving and forwards when comes in the 

transmission range of other nodes. These networks are 

also known as opportunistic because the node always 

searches for a connection opportunity to forward its 

buffered messages. The connection happens when

node(s) move in the communication range of each other. 

These transmissions continue and a message finds its 

destination via multiple hop(s). Based on this paradigm, 

various delay tolerant network (DTN) routing protocols 

have been proposed to cope with the heterogeneous 

network scenarios. The available routing strategies can 

be categorized as opportunistic, predicted and 

scheduled. The opportunistic protocol requires least 

computational resources and node simply forwards the 

message copy to the encounter nodes for example, 

Epidemic (Vahdat and Becker, 2000), Spray and wait 

(Thrasyvoulos et al., 2005). The predicted protocol 

forwards the message by utilizing the contact history of 

current node such as PRoPHET (Lindgren et al., 2003). 

While the scheduled based forwarding transmits the 

message by employing the human mobility movement 

pattern of contacts and positional coordinates, in DTN 

each node is hybrid, which means that it can generate 

and receive the random size messages. Hence, in the 

presence of limited bandwidth the small size messages 

swiftly pass through the network. This increases the 

message replication that results in rapid clogging on the 

buffer space of network nodes. Additionally, when a node 

by carrying small size messages receives the large size 

message it will drop the cluster of messages to 

accommodate new arrival. As, there exists multiple 

copies of each message, therefore same node may 

receive the dropped messages from other part of 

network. These rampant drop and re-transmissions 

wastes the bandwidth, energy and formulates high 

burden on the network resources. Hereby, a mechanism 

is required which control the generation of message 

copies and reduces the number of message drops. 

In this paper, we have proposed a new message 

routing strategy called Adaptive Message-Size routing 

strategy (AMRS),where a node handover the copy of 

message to its neighbors by employing a metric Mean 

Threshold (MTH). According to AMRS, before trans- 

mitting the message ‘M’ to ‘C’, the node ‘A’ computes the 

mean of buffered messages at ‘C’. If the size of message 

‘M’ is less than MTH of ‘C’ and ‘M’ is not the destination 

for ‘C’ then the ‘A’ will keep such messages in its buffer. 

However, ‘A’ will forward ‘M’ to ‘C’ only if message is 

destine for ‘C’ or the size of message is greater then 

MTH of ‘C’. 

We have compared the performance of AMRS with 

Epidemic and PRoPHET Protocols. Under all simulation 

configurations the AMRS always reduce the transmission 

overhead, message drop and raises delivery probability. 

RELATED WORKS 

DTN routing protocols attempts to transfer the data over 

the unstable communication. Thus, transmission of 

multiple message copies along different paths makes 

sure that at least one copy will reach the destination. An 

extreme example of such policy is the Epidemic (Vahdat 

Ayub et al. 1799 

and Becker, 2000) routing. In this protocol, when two 

nodes meet, they exchange all messages which they do 

not have in common. In this way, message(s) rapidly 

diffuses in the network and can reach the destination via 

multiple paths. This increases the robustness, minimize 

the delivery delay but consumes high volume of network 

resources such as bandwidth, buffer space and node 

energy. Some recent work (Qaisar et al., 2011) has 

optimized the performance of Epidemic routing by 

changing the default transmission order FIFO. The spray 

and wait routing (Thrasyvoulos et al., 2005) reduces the 

consumption of network resources by minimizing the 

transmission of message copies. In this scheme, the 

source node forward the ‘N’ number of its message 

copies to the encounter nodes called relays. This is spray 

phase. If the destination is not found in spray phase, then 

each node wait and forwards the message directly to 

destination. This protocol delivers the message like 

Epidemic fashion while it minimizes the consumption of 

network resource like direct delivery. Spray and focus 

(Thrasyvoulos et al., 2007, 2008) is another deviation of 

spray and wait where the node starts by spraying the N 

number of message copies. If the message does not find 

the destination then the node hands over the message to 

a custodian with higher utility. The utility value determines 

capability of node to meet the message destination. 

The contribution in routing protocol continues and the 

researcher begins to route the message by exploiting the 

additional network parameters such as the mobility 

pattern, encounter history activity of node. For example, 

the PRoPHET routing protocol (Lindgren et al., 2003) 

utilizes the encounter history and age of last encounter. 

Moreover, it route a message by using the metric of 

delivery predictability. NECTAR (Etienne and de Oliveira, 

2009) protocols controls the transmission of messages by 

computing the neighborhood index, which depends on 

the encounter history. Srinivasa and Krishnamurthy 

(2009) proposed a distributed proximity-based 

communication protocol that forwards the message using 

‘conditional residual’ time metric. This metric uses the 

local knowledge of past contact to estimate remaining 

meeting time for pair of nodes. CREST protocol 

minimizes the end-to-end delay and increase the delivery 

probability as compared to other message forwarding 

strategies MEED and MED. 

Conditional Shortest Path protocol (CSPR) (Eyuphan et 

al., 2010) effectively employs the human mobility pattern 

and computes the conditional intermitting time, which is 

the average meeting time of two nodes in relation to 

meeting time with third node. The average intermitting 

time represents the link cost. The node forwards the 

message by employing conditional shortest path 

algorithm that takes conditional average intermitting time 

as parameter. CSPR minimize the end-to-end delay and 

increase the delivery probability as compared to shortest 

path based routing protocols. Guizhu et al. (2010) 

improves the performance of Binary spray and wait 

algorithm by introducing QoN. The QoN is measured in


Figure 1. Appetizer node. 

terms of mobility of node and is represented by an integer 

number which indicates how frequent one node 

encounter with other node in a given time interval. Higher 

value of encounters indicates the high QoN. Messages 

are forwarded to a node that has higher value of QoN. 

When a node leaves one copy, it then switches to direct 

transmission. QoNsW increase the delivery probability, 

minimize the delivery delay, overhead as compared to 

Epidemic and spray and wait. 

PROCS (Jathar and Gupta, 2010) minimize the 

replication of message by adopting a forwarding strategy, 

which study the movement pattern of contacts and their 

time sequence. In PROCS, each node maintains a 

probabilistic contact graph that represents the contact 

probability among nodes at various time intervals. The 

node route the message by selecting a contact with 

greedy path calculation. PROCS protocol achieves high 

message delivery ratio with no message replication 

compared to Epidemic and PRoPHET. Daowen et al. 

(2009) manages the routing computation locally at each 

and transmits the message by using hierarchical 

forwarding with cluster control mechanism. This protocol 

split the network into different clusters. A node called 

cluster head controls each cluster. The node first sends 

the request to forward the message towards cluster head. 

The cluster head analyze the message header and 

perform Inter-cluster or Intra cluster forwarding. CHRC 

reduce the delivery delay and improve the delivery 

probability as compare to Epidemic and SMART 

protocols. The only drawback of CRHC is that it needs to 

maintain the information regarding partial nodes of the 

network. 

The proposed Adaptive Message-Size forwarding 

strategy forwards a message without exploiting and 

relegating the complex computations. The idea is to 

deliver or relay the message, the mobility of node and 

observing the buffer of encountered terminal. For this, a 

metric Mean Threshold (MTH) have been defined which 

controls the creation of multiple copies of each message. 

AMRS routing Algorithm (adaptive message size 

forwarding strategy) 

Terminology 

In proposed AMRS, each node maintains the index of its 

buffered messages. The node with high volume index 

controls the communication and is known as appetizer 

node. The volume index is determined in terms of 

number of message carried by a node. When two nodes 

have the same volume index then the selection of 

appetizer is random. The nodes ready to communicate 

with appetizer are known as active nodes. A node is 

active if it is not busy in communicating with its 

neighbors. 

The appetizer node controls the flow of messages 

around the active connections. For this, appetizer fetches 

its buffered message one at a time and transmits it by 

verifying the Rule 1, 2. After the ‘appetizer’ completes its 

transmission, it changes its status to active node while 

the next high volume index node becomes the ‘appetizer’. 

Figure 1 represents ‘A’ as appetizer while ‘B’, ‘D’ and ‘E’ 

are the active nodes. 

METHODOLOGY 

Rule 1: “The appetizer ‘A’ will forward message ‘M’ to ‘B’ only 

if ‘B’ is destination of ‘M’ and ‘B’ has not previously received 

the same message.” 

The existing Epidemic protocol (Vahdat and Becker, 2000) is a 

multi copy routing strategy where on each encounter, the node 

perform the pairwise exchanges of messages. Theses nodes 

become the carrier and continue to infect the further encountering 

terminals. In this way, the message rapidly moves across the 

disconnected regions. 

Despite delivering the messages, this protocol leaves high 

overhead on the network resources for instance buffer space, node 

energy and bandwidth. The previous work (Vahdat and Becker, 

2000) has taken the buffer space, bandwidth as infinite resources 

that is not practical for the real time DTN applications. For example, 

when the bandwidth is limited, then the node may not fully 

exchange their buffered messages and may miss the transmission 

of those messages for which the current connected node is the final 

destination. 

The message delivery to the destination is very important as 

when a message reach the recipients as a final custodian then that 

node generates an acknowledgement message. This message 

informs the other nodes about the status of delivered message and 

operates as a sweeper to remove the message copies and stop the 

further replication. However, when messages doest not reach the 

destination, their replication continues in the network. 

Rule 1 of the algorithm ensures the delivery of more messages to 

their destinations. For this, the appetizer node scans its buffer one 

message at a time and validates rule 1. Hence, the transmission 

occurs only if ‘M’ is destined for current connected node. However, 

when the message is not destined for ‘B’ then Rule II invokes. 

Rule 2: “The appetizer will not forward ‘M’ to ‘B’ only when the 

size of message ‘M’ is less than the mean of buffered 

messages at MTH (B).” 

In DTN, each node is capable of generating or receiving random

for each ACN in communication range of APN 

{ MTH (ACN) = BU/Mn; 

for each message ‘M’ in APN 

{ 

if (Mdest==ACN) 

forward_message (M,ACN); 

else if (Msize < MTH (ACN)) 

Continue; 

else 

Forward_message (M,ACN); 

} 

} 

Figure 2. Algorithm (AMRS). 

size messages. Further, these messages reach their destination via 

multiple intermediate hops called relays. Since, message relay is 

directly proportional to the congestion. The high number of 

message relays increases the network traffic. While under the 

limited buffer space, the nodes could not accommodate the allincoming 

transmissions. This congestion results in dropping the 

previously stored messages. As, there exists multiple copies of 

each message therefore same node may receive the dropped 

messages from other part of the network. These rampant drops and 

re-transmissions waste the bandwidth, energy and formulates high 

burden on the network resources. 

One factor that can be used to control the congestion is to 

minimize the number of message relays. Rule 2 deals with 

controlling the unnecessary relay of messages. For this, before 

forwarding the message ‘M’, appetizer ‘A’ computes the mean of 

buffered messages at ‘B’ which is called Mean Threshold (MTH). 

The mean value indicates the average size of message carried by 

the active node. If the size of ‘M’ at ‘A’ is less than the MTH (B), 

then it shows that as compared to ‘B’ the current message ‘M’ at ‘A’ 

is small. In addition, ‘B’ is not the destination of ‘M’ therefore 

relaying the message to ‘B’ does not guarantee its eventual 

delivery. At this stage, both nodes are equally capable to meet with 

the destination. Therefore relaying the message is waste of node 

energy, bandwidth and buffer space. Thus, ’A’ switch to the greedy 

mode and carry such small size messages in its buffer. 

As, all network nodes has random size messages. Therefore, 

MTH value varies for each node. Hence, appetizer adaptively will 

not relay small size messages and the transmission occurs only 

when size of message ‘M’ is greater than the MTH of active node 

(Figure 2). 

Message exchange 

Case 1: Exchange of Messages Epidemic Protocol 

Consider a sample scenario in Figure 3 where nodes ‘A’ and ‘B’ has 

established the connectivity while SVA, SVB represent the buffered 

messages at A and B. Hence, SVA = { (m1, 39 s, X, 200 KB), (m2, 

53 s T, 150 KB) ,(m3, 46 s, U, 180 KB), (m4, 22 s, B, 230 KB), (m5, 

10 s, B, 153 KB) | (message, arrival time, destination, size)}, SVB={ 

(m50, 41 s, X, 150 KB), (m51, 32 s, Z, 410 KB), (m52, 78 s, M, 200 

KB), (m53, 15 s, Q, 240 KB), | (message, arrival time, destination, 

size)}. 

According to Epidemic routing protocol, ‘A’ forwards its summery 


Figure 3. Exchange of messages using Epidemic routing protocol. 

vector to ‘B’. Node ‘B’ then computes the summery vector request 

SVR by subtracting the SVA from its own SVB to get the messages 

not buffered at ‘B’. 

The node ‘A’ then forwards the required messages to ‘B’ by 

arranging them according to the arrival time that is, the message 

with high arrival time should be placed on the top of the queue such 

as SV(A-B) = {m2, m3, m1, m4, m5}. 

Since bandwidth is the limited resource therefore it is possible 

that ‘A’ may not be able to forward m4, m5 that are placed at the 

end of the forwarding queue. Both m4 and m5 were destined for the 

current connection ‘B’. Therefore, instead of delivering the 

messages, ‘A’ will continue the replication m4, m5 on the other part 

of the network. 

Case 2: Exchange of messages via Adaptive Message Size 

forwarding strategy 

Now we will consider the transmission of messages via adaptive 

message size forwarding strategy. Consider the snapshot 

mentioned in Figure 4; here, ‘A’ act as appetizer because it has 

high volume index compared to ‘B’. Recall that appetizer node fetch 

its buffered messages one by one and forward it to the current


Figure 4. Exchange of messages using AMRS. 

active node by Rule 1 and Rule 2. 

In present scenario, appetizer ‘A’ will fetch ‘m1’ from its buffer 

and validate it by Rule 1. As m1 is not destined for ‘B’, thus by Rule 

2 ‘A’ will obtain the mean of the buffered messages MTH at ‘B’ that 

is MTH (B) = 250 KB. Since, the size of message m1 is less than 

the MTH (B) thus ‘A’ will keep ‘m1’ in its buffer and move to the next 

message ‘m2’. Similarly, ‘m2’ is not destined for ‘B’. In addition the 

size of ‘m2’ is also less than mean threshold MTH(B) thus ‘A’ keep 

this message in the buffer and moves to the next message ‘m3’ that 

will not be transmitted because it does not satisfy the algorithm 

rules. Thus, appetizer ‘A’ iterate the next message ‘m4’ which is 

destined for current connected node ‘B’; therefore ‘A’ will forward 

this message to ‘B’ and iterate the next message ‘m5’ that is 

destined for ‘B’; and ‘A’ will transmit ‘m5’ to ‘B’. 

SIMULATION AND RESULTS 

Here we will compare the performance of proposed 

Adaptive message size forwarding strategy (AMRS) with 

Epidemic and PRoPHET routing protocols by varying the 

various simulation parameters. The evaluation of 

schemes have been analyzed under ONE [22] simulator. 

ONE is a discrete event simulator written in JAVA and 

have been massively used by various researchers to 

measure the statistics of disrupted store-carry-forward 

applications. 

Network settings 

The simulations were performed by considering a citybased 

environment consisting two groups of pedestrians 

(80 nodes), one group of car (40 nodes) and six trains. 

The pedestrians and cars are moving according to 

Shortest Path Map Based Movement mobility model 

while trains are moving with Map Route Movement 

mobility model. The pedestrians were configured at the 

speed of 0.5 to 1.5 km/h for pedestrians, 10 to 50 km/h 

for cars and 10 to 50 km/h for trains with the transmission 

range of 10 m. We have configured each network peer 

with a finite buffer space and limited bandwidth 2 MBPS. 

In addition, the connection establishment is opportunistic 

and nodes do not have the knowledge about the network 

topology. 

Performance metrics 

Message relays 

DTN is an opportunistic network where messages reach 

its destination via multiple intermediate hop(s) referred to 

as relays. When a message is relayed, it consumes the 

energy, buffer space and bandwidth. The redundant 

diffusion of messages puts high load on them. Therefore, 

it is important to minimize the number of transmissions or 

relays. 

Message dropped 

Since, the transmission of multiple message copies 

produces high congestion on the buffer of intermediate 

node(s), in result, the node overcomes it by dropping 

previously stored messages. It is not possible to remove 

the drop event; however reducing its magnitude can 

improve the network throughput. 

Delivery probability 

The delivery probability measures the successful 

transmission of messages to their destinations. This 

metric measures the overall network throughput, as more 

messages delivery at destination shows the optimal use 

of network resources. 

Performance evaluation 

Scenario-01: Impact of varying number of nodes on 

routing protocols 

Figure 5 represents the results of PRoPHET, Epidemic

Figure 5. Message relays by varying number of nodes. 

Figure 6. Message drop by varying number of nodes. 

and AMRS in term of message relay. We can see that 

Epidemic protocol has the highest number of relays. The 

reason is that the message exchange rate of Epidemic 

protocol depends on the number of encounter. Thus, at 

higher nodes density such as 246, 216 and 186 the 

encounter rate among nodes also increases which in turn 

increase the maximum transmissions. 

The PRoPHET protocol has relayed fewer messages 

than Epidemic protocol. The reason is that PRoPHET 

Protocol does not relay the message if the encountered 

node is less probable to meet the destination. However, 

when we increase the node density for example [186], 

[216], [246] the PRoPHET protocol begins to relay more 


messages. The reason is that the PRoPHET protocol 

uses the history of encounters to control the replication of 

messages, in which high density of nodes elevates the 

encounters rate. In response more nodes has maintained 

the predictability measure for each other. 

The proposed AMRS shows the consistent message 

transmissions. The reason is that, AMPR generates the 

message copy only if the size of current message is 

greater than the MTH of connected node. In this way all 

nodes carry the small size messages and tries to deliver 

it directly to destination. 

Figure 6 shows the results of message drop in term of 

number of nodes. We can observe that the Epidemic


Figure 7. Delivery probability by varying number of nodes. 

protocol has dropped the high number of massages. The 

reason is that the Epidemic protocol blindly floods the 

message copies. While, under limited buffer space the 

encountering terminals could not accommodate all 

incoming traffic. 

The PRoPHET Protocol controls the replication of 

messages; therefore has dropped less messages than 

Epidemic. The AMRS reduces the message drop as 

compared to Epidemic and PRoPHET routing protocol. 

The reason for this is that, the Mean Threshold (MTH) 

moves the traffic to the less congested part of the 

network. 

Figure 7 depicts the result of delivery ratio for AMRS, 

PRoPHET and Epidemic routing protocols with respect to 

the number of nodes. We can see that the Epidemic 

protocol delivered the least number of messages. The 

reason is that, Epidemic protocol does not control the 

replication of messages as a result produce congestion 

on the network. Added together, protocol also overcomes 

this problem by dropping its carried messages. 

Therefore, high number of messages dropped before 

finding the destination. The PRoPHET protocol delivers 

more messages than Epidemic protocol. The reason is 

that PRoPHET protocol controls the replication of 

messages as well as forwards the message by 

computing the delivery probability. 

The proposed AMPR routing protocol double the 

delivery of messages for all simulation instances. 

However, at 246, the protocol has delivered maximum 

number of messages. The reason is that AMRS does not 

generate and forwards the copy of a small size message. 

This in turn, minimizes congestion, less number of drops 

and increase the message stay time in the buffer, which 

results in high delivery probability. 

Scenario 2: Impact of varying buffer size on routing 

protocols 

Figure 8 maps the results of AMRS, PRoPHET and 

Epidemic routing protocols in terms of message relays. 

We can observe that, as we increase the storage 

capacity such as 4 M, 5 M, 6 M, existing Epidemic and 

PRoPHET Protocol has increased the message relays. 

This is because message relay depends on the storage 

capacity of encounter node. 

The proposed (AMRS) routing strategy has maintained 

a consistent number of relays. The reason is that AMRS 

relays the message copy by computing the MTH and thus 

does not depend on the buffer size. 

Figure 9 shows the result of message drop for existing 

PRoPHET, Epidemic and proposed AMRS. We can see 

that for all simulation instances Epidemic protocol has 

high drop ratio. Moreover, as the storage capacity rises 

such as 4 M, 5 M, 6 M the drop also increases. The 

reason is that, Epidemic protocol relays high volume of 

messages that results in congestion. Thus, each node 

iteratively drops the previously stored messages to 

overcome the congestion and continue the flow of 

network traffic. The PRoPHET routing protocol controls 

the replication of message thus drop less messages than 

Epidemic. However, the proposed AMRS has dropped 

least amount of messages. The reason is that, the 

proposed strategy (AMRS) has minimum number of 

message relays. 

Figure 10 depicts the results of existing PRoPHET, 

Epidemic and proposed AMRS routing protocols in terms 

of delivery probability. We can see that with increase in 

the buffer size for example 3 M, 4 M, 5 M, and 6 M all 

routers has performed better. Moreover, AMRS routing

Figure 8. Message relay by varying buffer size. 

Figure 9. Message drop by varying buffer size. 

protocol has reciprocated well for all simulation instances 

and delivers messages two times higher than the existing 

strategies. The reason is that the delivery of message 

depends on the buffer time of the message; a message 

with high buffer stay time is more likely to be delivered to 

its destination. 

Scenario 3: Impact of varying message size on 

routing protocols 

Figure 11 represents the results of number of message 

relays for existing PRoPHET, Epidemic and proposed 

AMRS routing protocols by varying the sizes. We can 


evict that at the range of small size messages for 

instance [100 K – 600 K], [200 K – 700 K] both Epidemic 

and PRoPHET has the maximum number of 

transmissions. The reason is that the small size 

messages can float around the network more quickly. In 

addition, we can see that at large size messages for 

example [300 K -800 K], [400 K – 900 K] and [500 K – 1 

MB] all routing protocols has reduced the message 

relays. 

The proposed AMRS does not depend on the range of 

message size. We can observe that for all simulation 

instances the proposed protocol has shown a constant 

number of message relays. 

Figure 12 maps the finding of message drop by varying


Figure 10. Delivery probability by varying buffer size. 

Figure 11. Message relay by message size. 

the size of messages. We can observe that at small 

range of message sizes such as [100 K – 400 K], [200 K 

– 700 K], [300 K – 800 K] both Epidemic and PRoPHET 

has higher number of message drops. The reason is that 

small size messages tends to infect the network more 

rapidly and produces the congestion. 

Thus, nodes begin to drop the previously stored 

messages. In contrast, at large range for example [400 K 

– 900 K], [500 K – 1 MB] fewer messages were dropped. 

However, in case of AMRS, each node tends to carry the 

small size messages. 

Figure 13 conclude the ratio of successful transmission 

of messages to their destinations under the variable size 

of messages. In case of existing Epidemic and PRoPHET 

routing protocols, when the message range is small in 

size for example [100 K - 600K], [200 K – 700 K] more 

messages reaches the destination. The reason is that the 

small size messages are likely to find destination more 

quickly. However, the proposed routing protocol AMRS 

has delivered high number of messages to their 

destinations. The protocol has shown better results when 

there is high congestion in the network and message size 

is very small. 

Conclusion 

In this paper, we propose a new routing strategy known 

as adaptive message-size forwarding strategy (AMRS). 

According to the proposed strategy, a node generates

Figure 12. Message drop by random message size. 

Figure 13. Delivery probability by random message size. 

and handover the copy of message to neighboring nodes 

by using a metric named as Mean Threshold (MTH). We 

compared the performance of AMRS with Epidemic and 

PRoPHET Protocols. The proposed strategy has 

maximized the delivery probability while it controls the 

message drops and number of transmissions. 

REFERENCES 

Ahmedy I, Asri NMD, Syaril NOM, Junaid C (2011). A review on 

wireless sensor networks routing protocol: Challenge in energy 

perspective. Sci. Res. Essays, 6(26): 5628-5649. 

Daowen Hua, Xuehui Du, Yanbin Qian, Shaoge Yan (2009). Routing 

Protocol Based on Hierarchy Forwarding and Cluster Control. CIS’, 

pp. 397-401. 


Etienne C, de Oliveira R, C´ elio V, de Albuquerque N (2009). NECTAR: 

A DTN Routing Protocol Based on Neighborhood Contact History. SAC, 

pp. 8-12. 

Eyuphan B, Sabin CG, Boleslaw KS (2010).Conditional Shortest Path 

Routing in Delay Tolerant Networks. WoWMoM’, pp.1-6. 

Fall K, Farrell S (2008). DTN: an architectural retrospective. IEEE J. 

SelecT. Areas Commun., 5(26): 828-836. 

Guizhu W, Bingting W, Yongzhi G (2010). Dynamic Spray and Wait 

Routing algorithm with Quality of Node in Delay Tolerant Network. 

Jathar R, Gupta A (2010). Probabilistic Routing using Contact 

Sequencing in Delay Tolerant Networks.COMSNETS, pp. 1-10. 

Johnson DB, David A, Maltz J, Broch DSR (2001). The Dynamic Source 

Routing Protocol for Multi-Hop Wireless Ad Hoc Networks. Ad hoc. 

network. chapt., 5: 139-172. 

Lindgren A, Doria A, Schelen O (2003). Probabilistic routing in 

intermittently connected networks. SIGMOBILE Mobile Comput. 

Commun. Rev., 7(3). 

Qaisar A, Sulma RMSMZ (2011). TMHF: Transmit Max Hop First


forwarding Strategy to Optimize the Performance of Epidemic 

Routing Protocol. Int. J. Comput. Appl., 18(5): 40-45. 

Srinivasa S, Krishnamurthi S (2009). CREAST: An opportunistic 

forwarding protocol based on conditional residual time. SECON’09. 

pp. 1-9. 

Thrasyvoulos S, Konstantinos P, Cauligi SR (2005). Spray and wait: an 

efficient routing scheme for intermittently connected mobile networks. 

WDTN, 

Thrasyvoulos S, Konstantinos P, Cauligi SR (2007). Spray and Focus: 

Efficient Mobility-Assisted Routing for Heterogeneous and Correlated 

Mobility. PERCOMW, 

Thrasyvoulos S, Konstantinos P, Cauligi SR (2008). Efficient routing in 

intermittently connected mobile networks: the multiple-copy case. 

IEEE/ACM Trans. Network., 16(1): 77-90. 

Vahdat A, Becker D (2000). Epidemic routing for partially connected ad 

hoc networks. Duke University, Tech. Rep.



DOI: 10.5897/SRE12.255 



Length-weight relationships, relative condition factor 

and relative weight of three fish species from beach 

seine fishing grounds in Iranian coastal waters of 

Caspian Sea 

Moradinasab, Gh. 1 *, Raeisi, H 1 , Paighambari, S.Y. 1 , Ghorbani, R 1 and Bibak, Z. 1 

Department of Fisheries, Faculty of Fisheries and Environment, Gorgan University of Agricultural Sciences and Natural 

Resources, Gorgan, Iran. 


The aim of this study is to record the length-weight relationship, relative condition factor (Krel) and 

relative weight (Wr) for three fish species in Iranian coastal waters of the Caspian Sea. Fish sampling 

was carried out in the beach seine fishing grounds in autumn and winter seasons for two years (2007 

and 2009). 14104 specimens were measured and weighed. The values of the exponent b in the lengthweight 

relationships (LWRs) were 2.8449 for Cyprinus carpio, 2.8844 for Liza aurata and 2.9077 for 

Rutilus frisii kutum. Relative condition factor (Krel) values were ranged from 1.017±0.002 to 1.071±0.002. 

In addition, Relative weight (Wr) ranged from 0.929±0.002 to 1.740±0.004. 

Key words: Length-weight relationship, relative condition factor, relative weight, Caspian Sea. 

INTRODUCTION 

The Caspian Sea is the largest lake in the world that Iran 

is associated with that via the coasts of Guilan, 

Mazandaran and Golestan provinces (Alizadeh, 2004). 

This lake has a rich diversity of aquatic species. One of 

the common methods of fishing in this lake is the beach 

seine fishing which accounts for about 60% of total 

fishing in the area (Iranian Fisheries Statistic Yearbook, 

2008). Rutilus frisii kutum (Kamensky, 1901), Liza aurata 

(Risso, 1810) and Cyprinus carpio (Linnaeus, 1758) are 

the most important fish species in fishing composition 

and they account for more than 95% of the beach seine 

fishing. 

In terms of economic, employment and income, the 

three species mentioned before play a significant role in 

people's lives and livelihoods (Abdolmalaki and 

Ghaninejad, 2005). R. frisii kutum is the most valuable 

species in this fishing method and it is widely reproduced 

*Corresponding author. E-mail: Moradinasab88@yahoo.com. 

Tel: +98 911 7098 500. Fax: +98 0171 2245886. 

artificially in reproduction and breeding farms in Iran, and 

release into the Caspian Sea every year (Razavi sayad, 

1999). This species lives only in the basin southern 

Caspian Sea and so far, it has not been reported its 

length-weight relationships in Fish base and neither the 

other two species (L. aurata and C. carpio) in Iran 

(Froese and Pauly, 2012). 

In this study, length-weight relationship (LWR), relative 

condition factor (Krel) and relative weight (Wr) for these 

three species is discussed. 


Data collection 

Fish specimens were collected monthly from 130 fishing grounds 

beach seine with a net mesh size of 28 to 33 mm STR (48˚ 52′ E, 

38˚ 19′ N to 53˚ 45′ E, 36˚ 25′ N, Iran) in autumn and winter 

seasons for two years (2007 and 2009). There was no sampling in, 

spring and summer because they are close seasons for beach 

seine fishing in Iran. Numbers of different sizes were separated of 

combination fishing randomly and then the fresh specimens 

measured and weighed using the total length (TL, nearest 0.1 cm)


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. 

Species n 

Length (cm) WLR parameters and statistics 

r 2 

Mean S.E Min Max a SE (a) 95% CL(a) b SE (b) 95%CL (b) 

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 

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 

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 

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. 

and body weight (BW, nearest 1 g), respectively. Total data 

were pooled together in each species without sexing. 

Data analysis 

The length-weight relationships were estimated by using 

following equation (Froese, 2006): 

W = a L b 

Where W is the body wet weight (g), L is the total length 

(cm), a is the intercept of the regression and b is the 

regression coefficient (slope). The parameters a and b of 

the length-weight relationships were estimated by the 

least-squares method based on logarithms: 

Log (W) = log (a) + b log (L) 

A t-test was used for comparison the b values obtained 

from the linear regressions with isometric values (Sokal 

and Rohlf, 1987): 

Where ts is the t-test value, b the slope and Sb the standard 

error of the slope (b). The comparison between the 

obtained values of t-test and the respective tabled critical 

values allowed the determination of the b values 

statistically significant, and their inclusion in an isometric 

(b=3) or allometric range (negative allometric; b3). 

In addition, for each individual, the relative condition 

factor (Krel) and the relative weight (Wr) were calculated by 

following the equations (Le cren, 1951; Froese, 2006; 

Wege and Anderson, 1978): 

Where ts is the t-test value, b the slope and Sb the standard 

error of the slope (b). The comparison between the 

obtained values of t-test and the respective tabled critical 

values allowed the determination of the b values 

statistically significant, and their inclusion in an isometric 

(b=3) or allometric range (negative allometric; b3). 

In addition, for each individual, the relative condition 

factor (Krel) and the relative weight (Wr) were calculated by 

following the equations (Le cren, 1951; Froese, 2006; 

Wege and Anderson, 1978): 

Where W is the body wet weight (g), L is the total length 

(cm), a and b are the parameters of length-weight 

relationships, Ws is a standard weight representing the75th 

percentile of observed weights at that length, am is 

geometric mean a and bm is geometric mean b. Statistics 

were performed using the R software version 2.11.0. 

RESULTS AND DISCUSSION 

A total of 14104 specimens of three fish species 

were collected from the fishing grounds beach 

seine Iranian coastal waters of Caspian Sea 

during the present study. The number of 

individuals sampled (n), the length and weight 

ranges, parameters a and b of the length-weight 

relationships, the standard error of b value and 

the determination coefficient (r 2 ) for the three 

species are given in Table 1. 

In this study, relative weight was obtained for R. 

frisii kutum 0.929±0.002. Several factors influence 

in the growth of the fish such as hereditary 

characteristics, food reserves, environmental 

factors, pollution, etc. Investigations carried out 

indicate that growth R. frisii kutum in the recent 

years has decreased (Abdolmalaki and 

Ghaninejad, 2005). This can be mainly attributed 

to artificial breeding and restocking programs 

carried out every year by the Iranian Fisheries 

Organization (Table 2). 

Over fishing and the elimination of larger 

individuals of this species due to the use of 

inappropriate fishing gears have also contributed 

to this situation. Considering that there is no 

selection involved in choosing male and female 

spawners for artificial breeding programs, the

Moradinasab et al. 1811 

Table 2. Relative condition factor (Krel) (±S.E) and Relative weight (Wr) (±S.E) for the three fish species in the fishing 

grounds beach seine Iranian coastal waters of Caspian Sea. 

Relative condition factor (Krel) Relative weight (Wr) 

Species Min Max Mean (S.E) Min Max Mean (S.E) 

Rutilus frisii kutum 0.43 2.57 1.017±0.002 0.38 2.32 0.929±0.002 

Cyprinus carpio 0.45 2.91 1.029±0.004 0.69 4.30 1.594±0.006 

Liza aurata 0.38 3.28 1.071±0.002 0.62 5.25 1.740±0.004 

Min, minimum; Max, maximum; S.E, standard error. 

Table 3. Length-weight relationships obtained from other parts of the world for L. aurata and C. carpio. 

Species Location and references 

Liza aurata 

Cyprinus carpio 

Type of 

length 

Length 

(cm) 

Sex a b 

Portugal; Algarve (Borges et al., 2003) TL 20.1 - 40.5 Unsexed 0.0078 3.006 

Croatia; Eastern Adriatic, (Dulcic and Glamuzina, 2006) TL 21.5 - 44.2 Unsexed 0.0181 2.952 

Greece; G. Saronikos, (Stergiou and Moutopoulos, 2001) SL 15.5 - 21.0 Mixed 0.0078 3.230 

Japan; Shioda Plain, Nagano Prefecture, (Carlander, 1969) TL 31.5 - 57.0 Unsexed 0.0060 3.210 

Turkey; Lake Iznik, Marmara, (Tarkan et al., 2006) TL 14.2 - 48.8 Unsexed 0.0250 2.830 

Spain; Araquil River, Navarra, (Miranda et al., 2006) TL 7.1 - 59.0 Mixed 0.0120 3.070 

gene bank of this species is gradually shifting. 

Thus providing the required conditions for natural 

spawning in rivers where this species migrates to and or 

the semi-natural breeding of R. frisii kutum in earthen 

ponds where the selection of breeders is allowed even if 

only of a small size is stressed more than before 

(Khanipour and Valipour, 2009). 

A result of present study isn’t in agreement with reports 

of Stergiou and Moutopoulos (2001) and Carlander 

(1969) and is in agreement with reports of Tarkan et al 

(2006) (Table 3). 

In studies of population dynamics high condition factor 

values shows of favorable environmental conditions 

(such as: habitat and prey availability) and low values 

indicate less favorable environmental conditions 

(Blackwell et al., 2000). Relative condition factor (Krel) is 

commonly factor for indicate the condition of fish species. 

In our study, Liza aurata (1.071±0.002) had best 

performance, while Krel value in Rutilus frisii kutum 

(1.017±0.002) was lowest across caught species (Table 

2). 

Table 3 indicates a and b parameters of weight-length 

relationships of selected species obtained from other 

parts of the world. Our results mostly agreed with the 

sturgeon species studies given in Table 3. The difference 

of a and b can be affected area, sex, season, degree of 

stomach fullness, gonad maturity, health, habitat, 

nutrition (Tesch, 1971). 


We thank Iranian Fisheries Organization and Beach 

seine fishing cooperative societies for cooperation. 

REFERENCES 

Abdolmalaki Sh, Ghaninejad D (2005). Report on stock assessment and 

composition of the commercial bony fishes on the southern Caspian 

Sea. Fisheries Research Institute. Bandar Anzali, p. 132. 

Alizadeh H (2004). Introduction to the Features of the Caspian Sea. 

Norbakhsh publ., p. 120. 

Blackwell BG, Brown ML, Willis DW (2000). Relative Weight (Wr) Status 

and Current Use in Fisheries Assessment and Management. Rev. 

Fish. Sci., 8: 1-44. 

Borges TC, Olim S, Erzini K (2003). Weight-length relationship for fish 

species discarded in commercial fisheries of the Algarve (southern 

Portugal). J. Appl. Ichthyol., 19(6): 394-396. 

Carlander KD (1969). Handbook of freshwater fishery biology. Volume 

1. The Iowa State University Press. Ames. Iowa. DOI 

http://dx.doi.org/. 

Dulcic J, Glamuzina B (2006). Length-weight relationships for selected 

fish species from three eastern Adriatic estuarine systems (Croatia). 

J. Appl. Ichthyol., 22: 254-256. 

Froese R (2006). Cube law, condition factor and Length-Weight 

relationships: history, meta-analysis and recommendations. J. Appl. 

Ichthyol., 22: 241-253. 

Froese R, Pauly D (2012). Fish base. World Wide Web Electronic 

Publication. Available at http://www.fishbase.org. (accessed on 16 

February 2012). 

Khanipour AA, Valipour A (2009). Kutum jewel of the Caspian Sea. 

Iranian Fisheries Research Organization. Tehran. Iran, p. 97. 

Iranian Fisheries Statistic Yearbook (2008). Statistic catch fishes the 

basin southern Caspian Sea. Budget and code office- Statistic group 

and fishing developing study. Tehran Iran, p. 56. 

Le cren ED (1951). The length-weight relationships and seasonal cycle 

in gonad weight and condition in the perch (Perca fluviatilis). Anim. 

Ecol., 20: 201-219. 

Miranda E, Oscoz J, Leunda PM, Escala MC (2006). Weight-length 

relationships of cyprinid fishes of the Iberian Peninsula. J. Appl. 

Ichthyol., 22: 297-298. 

Razavi sayad B (1999). Introduction to the ecology of the Caspian Sea.


Iranian Fisheries Research Organization (IFRO). p. 90. 

Sokal RR, Rolf FJ (1987). Introduction to Biostatistics. 2nd Edition. 

Freeman. New, York, p. 363. 

Stergiou KI, Moutopoulos DK (2001). A review of length-weight 

relationships of fishes from Greek marine waters. Naga ICLARM Q. 

24(1 and 2): 23-39. 

Tarkan AS, Gaygusuz O, Acipinar H, Gürsoy C, Ozulug M (2006). 

Length-weight relationships of fishes from the Marmara region (NW- 

Turkey). J. Appl. Ichthyol., 22: 271-273. 

Tesch FW (1971). Age and growth. In: Ricker, W. E. ed., Methods for 

Assessment of Fish Production in Freshwaters. Blackwell Scientific 

Publications. Oxford, pp. 98-100. 

Wege GJ, Anderson RO (1978). Relative weight (Wr): A new index of 

condition of largemouth bass. In: New approaches to management of 

small impoundments. G. Novinger J. Dillard (Eds). Am. Fish. Soc. 

Spec. Publ.. Bethesda. MD, 5: 79-91.



DOI: 10.5897/SRE11.1815 



Design and development of standalone DSP prototype 

for QT interval processing and monitoring 

Goh Chun Seng*, Sh-Hussain Salleh, J. M. Najeb, I. Kamarulafizam, Mahyar Hamedi and Alias Md Noor 

Centre for Biomedical Engineering (CBE), Transportation Research Alliance, Universiti Teknologi Malaysia, 813010 

Skudai, Johor, Malaysia. 

Accepted 30 March, 2012 

This paper describes the development of stand alone DSP hardware for QT interval monitoring and 

assessment. The QT interval has been known to be an important indicator prior to Myocardial Infarction 

(MI) and it is important to observe and monitor any changes in the period of the QT interval. The system 

consists of three units. The first unit includes floating point digital signal processor, TMS320VC33 which 

is selected for the development of signal processing with memory circuit. The following unit includes the 

development of signal monitoring displays circuit with universal serial bus (USB) interface. The third 

unit includes the development of analog front end (AFE) circuit and merged with signal conditioning 

circuit. This paper discusses the principle of system design and the system has the advantages to 

maximize utilization by allowing modification, reconfiguration, portable, cost effective and run in standalone 

operation. As well as eligibility of on board pre-program algorithm for QT analysis which existing 

ECG monitoring equipment are lacking. The system had been successfully tested with algorithm for QT 

analysis and it is validated with healthy subject’s data files from PTB diagnostic ECG data base. 

Key words: QT interval, DSP processor, stand alone operation, portable. 

INTRODUCTION 

The invention of ECG in the beginning of 20 th centuries 

by Willem Einthoven remains as demanded tool and 

popular component until today for non-invasive method to 

evaluate heart condition of patients who have been 

diagnose with signs and symptoms in clinical environment 

(Sandham et al., 2007). The interpretation of ECG 

will allow cardiologist and clinicians to understand and 

collect more information of electrical activity of the heart. 

However, with the integration of embedded technology, 

the automated interval measurement of P,Q,R,S and T in 

the ECG recording system has became an ideal 

instrument for patient monitoring and supervision in 

clinical practice. Although, the automated measurement 

are well establish in the ECG recording system but 

recently (Kligfield et al., 2006) there are studies focuses 

*Corresponding author. E-mail: seng4@yahoo.com. Tel: 07- 

5535933. 

on algorithm variation of automated QT interval 

measurement on commercial ECG recording system from 

different manufacturer. 

However, the automated methods of the QT interval 

are already widely available for computerized electrocardiography 

system (Christov, 2006), which ranges from 

manufactured ECG recording system to PC based 

system with different size and weight. Nevertheless, 

those commercial systems of automated QT interval 

measurements from different manufacturers have been 

utilize in studies to predict heart concern, (Kligfield et al., 

2006) but due to different automated QT interval in the 

system with different automated algorithm from different 

manufacturers which can affect the identification and 

quantification of the acquired ECG, which means that 

different automated of QT measurement algorithm 

produces variation and error when validated against 

manually annotated QT database. 

The QT interval measurement is crucial and it is used 

as an informative index to predict sudden death due to


cardiac arrhythmias (Malarvili et al., 2005). The indication 

of cardiac arrhythmias is related to the duration of the QT 

interval and it is important to observe and monitor any 

changes in the period of the QT interval (Mneimneh et al., 

2006). 

Traditionally, QT interval measurements of patients are 

interpreted manually from ECG strip paper printed by 

ECG recording machines, the measurement are 

dependent and costly, if require skillful and experienced 

cardiologist or physician to carry out the manual 

procedures (Malik, 2004). Therefore the Food and Drug 

Administration (FDA) has initiated the encouragement to 

recommend digitizes ECGs and it is highlighted by 

Badilini (2005) as an efficient way to triggered the interest 

toward algorithm based ECG analysis. 

Given the fact that, commercial automated QT interval 

measurement with signal analysis capabilities is an 

expensive customize systems (Oweis and Hijazi, 2006), 

the reason are, the first; systems development timeline 

are strictly bounded by procedures from regulator. 

Second, it is intended to be use in clinical practice for 

human subject only (Saviola, 2005). Thus, only authorize 

manufacturer is given the mandate by regulator to make 

modifications and improvements toward improving speed 

and efficiency of the commercial system (Kaplan et al., 

2004). Therefore, that restriction by regulator will be a 

financial challenge for researcher’s to acquire various 

types of instruments with QT interval measurement for 

just to harvest the raw ECG from patient and do offline 

QT analysis on computer. It is clearly to shows that, the 

commercial automated QT interval measurement 

systems are not subject for any research purposes in 

term of reprogrammable the QT algorithm on those 

commercial systems for ECG data analysis. 

Therefore, in this paper we introduce the development 

of an automated standalone system which is focused on 

portability, standalone operation and low cost QT interval 

measurement device. This standalone system is using 

Digital Signal Processor (DSPs) technology to 

accomplish the real time QT interval measurement from 

the tapped ECG and it is reprogrammable with simplest 

to more complex QT algorithms; ranges begin with simple 

beat to beat detection followed by advanced diagnosis for 

disorder heart disease. There are various systems 

applications in ECG parameters analysis have been 

reported (Mohammad et al., 2003; Gorup et al., 2000; 

Kara et al., 2006) but none of them describes a real time 

QT interval measurement regarding standalone operation, 

portability and low cost device. 

The high computational performance of Digital Signal 

Processor is sufficient to implement a standalone system 

without any host controller, such as computer. It is also a 

highly flexible standalone system and can be easily 

reprogrammed (Soon et al., 2001). Therefore, the 

advantages of the proposed standalone system, will allow 

cooperation between the researcher and cardiologist to 

work on implementation of QT algorithm development 

with on board reprogrammable function which is lacking 

in the existing commercial automated QT interval 

measurement systems. 

METHODOLOGY 

The architecture layout of the system 

The main components of the standalone system design will be 

described in stages according to the logically organization. The 

description will begin with Analog Front End Unit, follow by Signal 

Monitoring displays Unit, Signal Processing with Memory Unit and 

Power Supply Unit. All the stages of the process will be described. 

Figure 1 is the block diagram of the stand alone system and the 

way of components are interconnected and interrelated. 

Signal processing with memory unit 

This unit is focuses on incoming ECG from codec chip set for signal 

processing. There are few major components are used in the signal 

processing with memory unit. The mains components are 32 bit 

floating point DSP processor (TMS320VC33), Electrical Erasable 

Programmable Read only Memory - EEPROM (CAT28LV64W) and 

Static Random Access Memory - SRAM (CY7C1041DV33). The 

origin of DSP processor itself is design to handle mathematical and 

algorithm application and it is inefficient as a task oriented 

controller. For the SRAM memory in this unit is to allow for ECG 

data storage and the SRAM memory are expandable in both the 

data bus and address bus width. 

Analog Front End (AFE) Unit 

This unit provides interaction through peripherals connectivity with 

the real world environment. The peripherals are remote or 

pluggable analog front end devices either ready made or 

commercial or custom build ones to allow converted input signal to 

be delivered into the main board. Figure 1 shows that the tap ECG 

is delivered to the main board through the custom build AFE device. 

This AFE input consists of: (a) Signal conditioning circuitry (SCC), 

(b) Codec chip. 

The SCC is needed to amplify the tap ECG (0.5 to 5mV peak to 

peak) to at least the gain of 1000 (Najeb et al., 2005). Noises are 

common for tap ECG signal from body skin surface. Therefore, this 

SCC is build with instrumentation amplifier (INA121P) with low 

noise differential signal acquisition in order to minimize noise 

interference. The output from SCC is connected to codec chip set 

(PCM3003) to interface with TMS320VC33 in the main board. 

The codec chip is fully controlled by hardware setting and the 

data formats are 16 bit. Then the data transmission is only clock in 

and clock out in sequence from the codec chip set. So, it is need 

not depending on software setting or external clocking for 

operation. 

Signal monitoring unit 

This unit provides display on graphical liquid crystal displays 

(GLCD) for monitoring ECG as shown in Figure 1. In this unit, the 

microcontroller, PIC18LF452 is configured to interact with graphical 

liquid crystal displays (GLCD). The microcontroller is added in this 

unit without interrupting the TMS320VC33, because the 

microcontroller performs best in tasking order, which capable of

SCC 

Codec 

ADC 

Analog Front End unit 

Main Board 

Remote socket 

GLCD 

Microcontroller 

DSP 

RAM 

Signal monitoring 

displays unit 

Signal processing with 

Memory unit 

Seng et al. 1815 

USB 

ROM 

Figure 1. Block diagram of the stand alone system. The core processor is a floating point, Digital Signal Processor, 

TMS320VC33. The capture ECG signal is display on the Graphic LCD (128x64 dot pixel).l. 

task orientation and lack of DSP terms. However, the ADC0820 

received the input from the tap ECG signal through signal 

conditioning circuit (SCC) and the converted ADC data is latch to 

the output port of ADC0820. Then the tap ECG signal from SCC 

output was capture and converted by 8 bit Analog to Digital 

Converter chip set (ADC0820) and translated by PIC18LF452 into 

GLCD code in order to display on the GLCD. 

Figure 2, shows that, upon initialization, the microcontroller put to 

observe the FT245RL chip set to get acknowledge and the 

observation is done through Port A of PIC18LF452. As the 

acknowledge signal is detected, and then the firmware will proceed 

to next step’s label as ‘A’. If there is no feed back for acknowledge 

signal by the FT245RL, then in this case, the next stage will begin 

with monitoring the external interrupt and the firmware loops. If the 

interrupt is detected, PIC18LF452 will initialize the Graphic LCD 

panel and begin to read the ADC data from Port C of PIC18LF452. 

After that, the tap ECG is put on GLCD screen to display as the 

ECG monitoring signal. 

As depicted in Figure 3, there are two categories in the signal 

monitoring displays unit. The first category is the microcontroller 

with USB interfacing, colored with yellow dotted line. The FT245RL 

provides pin WR and pin RD# which are functions as write and read 

with 128 byte and 256 byte buffers in the FT245RL. The functions 

of the WR and the RD# pins are associated with pin TXE# and pin 

RXF# to indicate the buffer status, either it is empty so that written 

process is allowed to proceed or whether the buffer is full so that 

the read process can be executed. However, only pin WR and pin 

TXE# from FT245RL are utilized and both of these pins are 

connected to PIC18LF452. When TXE# is in LOW state means that 

the write buffer is ready and the data is awaiting at the FT245RL 

buffer. Then, the WR pin is toggled by PIC18LF452 from the HIGH 

state to LOW state, in order to transfer the data in bulks of 128 

bytes into the EEPROM. The EEPROM enable by setting both 

CE and WE pins are in LOW state and the OE pin is in HIGH 

stage. 

The lengths of the data in bulk mode are 21 bits and the transfer 

mode is in serial. Therefore, the 74HC164 with 8 parallel output is 

connected in cascade of 3 units in a row, so that this connection will 

allow the 21 bits of data to transfer as serial in and parallel out from 

the 74HC164. Figure 4 shows the circuit for microcontroller with 

USB interfacing and serial in parallel out circuit in signal monitoring 

displays unit. In Figure 5 shows the firmware loops and 

downloading process for pre-program ECG algorithm into the 

EEPROM in serial packet of address and data via USB connectivity. 

The firmware begins with the initialization of EEPROM and the 

shift register. After that, the download begins with sequence of 21 

bit per packet in serial format. Hence, the 21 bit per packet is 

shifting into the EEPROM by three shift register -74HC164 cascade 

in serial with each shift a byte of address and data into the 

EEPROM. The format of 21 bit per packet of address and data is 

depicted in Figure 6. The shift register is connected with FT245RL 

and it is monitored by PIC18LF452 to manage the elapse period of 

data transfer and will received acknowledge after data transaction 

is completed. 

Power supply unit 

This prototype DSP hardware is designed to utilize the USB power. 

The focus is on low power consumptions. The USB voltage is 

regulated by TPS77533 and TPS77618 to provide necessary power 

demand in signal processing with memory unit, signal monitoring 

displays unit and analog front end (AFE) unit. Both of the voltage 

regulators are low drop out (LDP) voltage regulator. The ranges of 

different power supply are group as the following: 

i. 1.8V for DSP processor core supply. 

ii. 3.3V for DSP processor I/O pins and External Memory chip set. 

iii. 3.3V for Microcontroller. 

iv. 5.0V for Voltage regulator.


QT Algorithm for ECG Signal Analysis 

Start 

Initialize: 

- Port A, Port B, Port C and Port D 

No 

FT245RL, get 

acknowledge? 

No 

Interrupt, 

ADC0820 

Yes 

? 

Initialize: - GCLD 

Yes 

Read the ADC data from Port C 

Display the digitize signal to GLCD as 

monitoring signal 

End 

Figure 2. Flowchart of firmware loop to display the tap ECG to the GLCD. 

The QT algorithm applied in this prototype DSP hardware is 

adapted from Laguna P et al. (1990). The criteria’s of these QT 

algorithm involves the preprocessing, QRS detection, R and Qwaves 

definition, QRS onset definition, T wave peak and T wave 

end definition. The criteria’s are depicted in Figure 7. The first 

attempt, a low pass differentiator (LDP) and first order low pass 

filter is used to eliminate noise (the residual and intrinsic 

differentiation noise). Next stage is the QRS detection, by defining 

threshold (H1) to detect the QRS peak. After detecting the QRS 

peak, the search of nearest peaks begins with forward and 

backward search, in order to identify the R peak. After that, the R 

and Q position (Rp and Qp) are define through zero crossing 

method. Then, continue with QRS onset definition, where 

backwards search is carry out in the differentiated signal d(k) for Qi 

and Ri in order to identify the present of Q and R wave (Laguna et 

al. 1990). Next stage, the QRS begin (QRS1) is define from the zero 

crossing point of Qi and Ri. From the R position Rp, a search 

A 

windows are define, the bwind and ewind. Then the searching is 

beginning with forward search to look for maximum and minimum 

value, which will define the T wave end in the searching process. 

This method is selected because it is produces the closet results 

(Moraes and Viana, 1997) obtained visually by specialist when 

tested for single lead (lead II). The QT algorithm was developed in 

C language with the assistant of Code Composer 3x/4x before it 

was implemented in the prototype DSP hardware. Upon Validation 

on this prototype DSP hardware for the developed QT algorithm, an 

annotated data set of Physikalisch Technische Bundesanstalt 

(PTB) Diagnostic ECG data prepared Christov et al. (2006) is used. 

As stated by Najeb (2007), the QT algorithm works best on healthy 

subjects in PTB diagnostic ECG data base. Therefore, this is done 

by downloading the PTB Diagnostic ECG healthy subject’s files 

only, from the data base into the prototype DSP hardware. 

The implementation and validation of this QT algorithm to the 

prototype DSP hardware is described in the flow diagram as 

depicted in Figure 8. The experiment setup procedure is to validate 

the QT algorithm for the prototype DSP hardware. The process


Data / 

Address 

USB 

FT245RL 

EEPROM Shift Register 


TXE# 

WR 

CLK 

Data 

/WE 

/OE 

S.I.P.O 

Signal Monitoring Displays unit 

Figure 3. Block diagram of hardware interfacing in signal monitoring displays unit. 

CLR 

9 

CLK 

8 

A 

1 

B 

2 

QA 

3 

QB 

4 

QC 

5 

QD 

6 

QE 

10 

QF 

11 

QG 

12 

QH 

13 

VCC 

14 

GND 

7 

srU1 

CLR 

9 

CLK 

8 

A 

1 

B 

2 

QA 

3 

QB 

4 

QC 

5 

QD 

6 

QE 

10 

QF 

11 

QG 

12 

QH 

13 

VCC 

14 

GND 

7 

srU3 

srVCC 

srVCC 

srVCC 

srVCC 

srVCC 

srVCC 

srCLK_IN 

srCLK_IN 

srCLK_IN 

sr3D8 

sr2D8 

sr1D8 

sr1 D 8 

sr2 D 8 

CLR 

9 

CLK 

8 

A 

1 

B 

2 

QA 

3 

QB 

4 

QC 

5 

QD 

6 

QE 

10 

QF 

11 

QG 

12 

QH 

13 

VCC 

14 

GND 

7 

srU2 

1 2 

3 4 

5 6 

7 8 

9 10 

srJP1 

srVCC 

srD ata_ IN 

srCL K _ IN 

srD ata_ IN 

sr/W E 

sr/CE 

sr/O E 

A0 

10 

A1 

9 

A2 

8 

A3 

7 

A4 

6 

A5 

5 

A6 

4 

A7 

3 

A8 

25 

A9 

24 

A10 

21 

A11 

23 

A12 

2 

CE 

20 

OE 

22 

WE 

27 

GND 

14 

VCC 

28 

DQ0 

11 

DQ1 

12 

DQ2 

13 

DQ3 

15 

DQ4 

16 

DQ5 

17 

DQ6 

18 

DQ7 

19 

NC 

1 

NC 

26 

srU5 

sr/WE 

sr/CE 

sr/OE 

srVCC_S 

sr1D1 

sr1D2 

sr1D3 

sr1D4 

sr1D5 

sr1D6 

sr1D7 

sr1D8 

sr1D1 

sr1D2 

sr1D3 

sr1D4 

sr1D5 

sr1D6 

sr1D7 

sr2D1 

sr2D2 

sr2D3 

sr2D4 

sr2D5 

sr2D6 

sr2D7 

sr3D1 

sr3D2 

sr3D3 

sr3D4 

sr3D5 

sr3D6 

sr3D7 

sr2D8 

sr2D1 

sr2D2 

sr2D3 

sr2D4 

sr2D5 

sr2D6 

sr2D7 

sr3D1 

sr3D2 

sr3D3 

sr3D4 

sr3D5 

1 2 

3 4 

5 6 

srJP2 

srVCC_S 

srVCC_S 

w1 w1 w3 

w3 

vr5V ldo3V 

1 

2 

uY1 20Mhz 

uC7 

22p 

uC8 

22p 

uS1 

10K 

uR13 

uVP5 

470 

uR12 

RC7/RX/DT 

26 

RD4/PSP4 

27 

RD5/PSP5 

28 

RD6/PSP6 

29 

RD7/PSP7 

30 

VSS 

31 

VDD 

32 

RB0/INT0 

33 

RB1/INT1 

34 

RB2/INT2 

35 

RB3/CCP2* 

36 

RB4 

37 

RB5/PGM 

38 

RB6/PGC 

39 

RB7/PGD 

40 

RC6/TX/CK 

25 

RC5/SDO 

24 

RC4/SDI/SDA 

23 

RD3/PSP3 

22 

RD2/PSP2 

21 

RE2/^CS/AN7 

10 

RD1/PSP1 

20 

RD0/PSP0 

19 

RC3/SCK/SCL 

18 

RC2/CCP1 

17 

RC1/T1OSL/CCP2* 

16 

RC0/T1OSO/T1CKL 

15 

OSC2/CLKO/RA6 

14 

OSC1/CLKL 

13 

VSS 

12 

VDD 

11 

RE1/^WR/AN6 

9 

RE0/^RD/AN5 

8 

RA5/AN4/^SS/LVDIN 

7 

RA4/T0CKL 

6 

RA3/AN3/VREF+ 

5 

RA2/AN2/VREF- 

4 

RA1/AN1 

3 

RA0/AN0 

2 

/MCLR/VPP 

1 

uU2 

1 

2 

3 

4 

5 

6 

7 

8 

9 

10 

11 

12 

13 

14 

15 

16 

17 

18 

19 

20 

21 

22 

23 

24 

25 

26 

27 

28 

29 

30 

31 

32 

33 

34 

35 

36 

37 

38 

39 

40 

uJP5 

1 

2 

3 

4 

5 

6 

7 

8 

9 

10 

11 

12 

13 

14 

15 

16 

17 

18 

19 

20 

21 

22 

23 

24 

25 

26 

27 

28 

29 

30 

31 

32 

33 

34 

35 

36 

37 

38 

39 

40 

uJP4 

uVP5 

uVP5 

VBUS 

1 

D- 

2 

D+ 

3 

GND 

4 

uUSB_B1 

1 

2 

uJP3 

uVP5 

USBDM 

USBDP 

4K7 

uR5 

10K 

uR6 

4.7uF 

uC5 

104pF 

uC3 

104pF 

uC2 

uVP5 

104pF 

uC1 

uLD8 

uLD9 

uVP5 

1 

2 

3 

4 

5 

6 

7 

8 

9 

10 

11 

12 

13 

14 

15 

16 

17 

18 

19 

20 

21 

22 

23 

24 

25 

26 

27 

28 

uJP2 

D0 

1 

D4 

2 

D2 

3 

VCCIO 

4 

D1 

5 

D7 

6 

GND 

7 

NC 

8 

D5 

9 

D6 

10 

D3 

11 

PWREN# 

12 

RD# 

13 

WR 

14 

USBDP 

15 

USBDM 

16 

3V3OUT 

17 

GND 

18 

RESET# 

19 

VCC 

20 

GND 

21 

TXE# 

23 

RXF# 

22 

NC 

24 

AGND 

25 

TEST 

26 

OSCI 

27 

OSCO 

28 

uU1 

1 

2 

3 

4 

5 

6 

7 

8 

9 

10 

11 

12 

13 

14 

15 

16 

17 

18 

19 

20 

21 

22 

23 

24 

25 

26 

27 

28 

uJP1 

USBDM 

USBDP 

Figure 4. Serial in Parallel out (SIPO) circuit and EEPROM interface circuit.


A 

Initialize: 

- Shift Register and EEPROM 

Data Packet: 

- Address byte and Byte count 

Start Downloading to EEPROM 

Done? 

End 

Yes 

Figure 5. Flow chart of downloading data to EEPROM. 

Location 1 

Location 2 

Location 3 

1 st Address 

2 nd Address 

3 rd Address 

+ 

+ 

+ 

21 bit 

No 

1 st Data 

2 nd Data 

3 rd Data 

Location n n th Address n th + Data 

Figure 6. Transfer data packet of 21 bit to EEPROM via USB. 

begins with hardware setups followed by Code Composer to load 

the QT algorithm from the created workspace window and deliver 

each file (healthy subjects) from the PTB Diagnostic ECG data base 

by using probe point into the prototype DSP hardware. The data 

file format for PTB Diagnostic ECG data base is converted into the 

Code Composer 3x/4x format. As the processing begins, the signal 

transition is visible on the graphical display window as shown in 

Figure 9 for the DSP prototype hardware so that it enables 

researcher to observe and gain insight for each executed 

processing step. Finally, the obtained result for Q onset and T offset 

are displayed at the watch window for the prototype DSP hardware 

as depicted in Figure 9. The obtained results are validated with the 

annotated data set. 

The ECG data for validation uses in Code Composer 3x/4x is 

acquired from PTB diagnostic ECG database. In this ECG database 

only information of ECG lead II is extracted for QT interval 

measurement. However, the extracted lead II information are in 

numeric text format but the required format for File I/O using probe 

point in Code Composer 3x/4x, must be in hexadecimal. Therefore, 

conversion of file format is required and needed to be done in order 

for Code Composer 3x/4x to read those ECG files to the target 

processor, TMS320VC33. 

The requirements for the requested ECG files are in 2 bytes 

format for each data present in the converted file from the PTB 

Diagnostic ECG data base. 

Figure 10 shows a converted file of patient104_s0306lre taken 

from PTB Diagnostic ECG data base: On the left of Figure 10 

shows the raw PTB Diagnostic ECG data with file name:

Preprocessing 

- Low pass differentiator (LPD) 

- 1 st order low pass filter. 

QRS detection 

- Define Threshold (H1):QRSmax and QRS min 

- Identified R peak. 

R and Q waves definition 

- Define R and Q position (Rp and Qp) from 

zero crossing. 

QRS onset definition 

- QRS-begin is define from the zero 

crossing of Rp and Qp. 

T-wave peak and T-wave end definition 

- Define a search window from the Rp. 

- bwind and ewind, forward search. 

Figure 7. Simplified flow chart for Q onset and T offset 

detection. 

Start 

Setup DSP Hardware 

Initialize Code Composer 

Load workspace 

Load Program (QT algorithm) 

Compile and build 

Load “QT algorithm’’ onto DSP processor 

Set and Connect Probe Point to acquired downloaded files’ from PTB 

Diagnostic ECG database (Healthy subjects) 

Observe signal transition (Graphical Display Window) 

Display Q Onset and T Offset results (Watch Windows) 

Validate the results with annotated Q onset and T offset data set 

End 

Figure 8. Flow diagram of QT algorithm implementation and validation for the developed prototype DSP 

hardware. 

Seng et al. 1819


ECG file 

Watch window: Measured result for QT interval 

Signal transition 

Figure 9. The QT algorithm is developed in C with the assistant of Code Composer 3x/4x. The signal transition is 

visualized at the watch windows for each preprocessing interface to the prototype DSP hardware. 

Raw 

ECG Data, from PTB 

Diagnostic ECG 

data base 

Converted 

ECG data 

File Header 

Figure 10. The ECG file conversion from raw ECG data to 2 byte format, to be used with Code Composer 3x/4x.

Fix number 

Format Starting address Page number 

Samples data in the file 

Length 

Figure 11. File header information, the converted ECG file (patient104_s0306lre) with 2 byte format. 

patient104_s0306lre is extracted from Physio bank ATM for lead II 

information and on the right is the converted information, in 16 bit or 

2 byte format. The converted ECG file of patient104_s0306lre 

begins with a header format located in the first line of the files. 

The header format is depicted in Figure 11. The first line of the 

converted file is the header’s information which is in hexadecimal 

format. The header consists of useful and needed information for 

ECG file processing. The header information syntax is a) fix 

number, b) data format, c) starting address of memory block, d) 

page number e) length of block data. 

(a) Fix number: Fixed by default numeric number, 1651. 

(b) Data format: Is a numeric number beginning with 1 to 4, to 

indicate the samples format in the file. This numeric number 

represents a data format: hexadecimal, integer, long integer, 

floating point, respectively. However, numeric 1 is selected to 

indicate that the entire converted ECG file for patient104_s0306lre 

is in hexadecimal. 

(c) Starting address: To indicate the start address for data file. In 

this case the setting is 0. 

(d) Page number: Is a numeric number beginning with 1 to 2, to 

indicate the type of file contents. This numeric number represents: 

data and program respectively. However, numeric 1 is selected to 

indicate the file contents are only data. 

(e) Length: To indicate the total samples data in the file. In this 

case the setting is 0. 

Nevertheless, there are two numeric zero is defined in the file 

header as shown in Figure 11. These two numeric zeros are 

representing the starting address of data file and the length of data 


file. But, these two zeros can be overwritten, when using the Code 

Composer data file format with file I/O capabilities as depicted in 

Figure 12, this means that any information entered in the Address 

dialog box and in the Length dialog box, will automatically override 

the Code Composer data file information. Followed by, the 

information which is entered, to the dialog box, the address begins 

with inp_buffer and the lengths are 3200 samples of data. The 3200 

samples of data are the converted ECG file of patient104_s0306lre 

as shown in Figure 10. The address and the length of data also 

defined in C, respectively as depicted in Figure 12b. In order to link 

the file, as shows in Figure 12c it is needed to connect probe point 

with the converted ECG file of patient104_s0306lre. After that, is to 

execute the QT algorithm under Code Composer 3x/4x as depicted 

in Figure 12d. 

RESULTS 

Experiment had been carried out in order to determine 

the functionality and performance for this standalone 

DSP prototype design. The ECG parameter chosen for 

this experiment setup were Q onset and T offset. 

Therefore, this prototype was validated by using the PTB 

diagnostic ECG database utilizing the annotated QT data 

set prepared by Christov et al. (2006). An emulator test 

setup with code composer 3x/4x were used to test the 

standalone prototype DSP hardware functionally and


a) Setup address and length 

for input buffer 

Correction 

b) Address and buffer for the data length using 

the C file system in Code Composer 3x/4x 

c) A converted ECG file of patient104_s0306lre is connected with probe point. 

d) The QT algorithm read in 3200 word of data from a converted ECG file of patient104_s0306lre. 

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. 

performance. 

The test setup was organized into two parts: a) the 

functional test results and b) the validation result. In 

functional test results, tap ECG were acquired from AFE 

unit to be displayed on the graphical LCD (GLCD) display 

and the on chip (TMS320VC33) physical pin H1 is 

checked to ensure the present of signal oscillation in 

multiple of scale five factors from the clock input 

(oscillator). For validation results, ECG file from PTB 

diagnostic ECG database were used and compared with 

the annotated data set, the obtained results were plot in 

line graph. The functional test and validation results will 

be discussed. 

Functional test results 

When power on, pin H1 was checked. The result is

100MHz 

20MHz 

Figure 13. Pin H1 at TMS320VC33, capture by scope-meter software (fluke, model-199b). 

shown in Figure 13, the 100MHz is the multiple x5 scale 

factor from the clock input which is the oscillator 

frequency, 20MHz. This proves that the chip set 

(TMS320VC33) is operational. After that, ECG signal is 

acquired through the input from AFE as shown Figure 14. 

The AFE hardware is custom build and the ECG signal is 

obtain from the Lead II formation on human body skin 

surface with fours disposable electrodes place on the 

right hand, left hand, right leg and left leg. 

This custom build AFE hardware device consists of: a) 

Signal Conditioning Circuitry (SCC) and b) codec chip, 

PCM3003. The SCC is needed to amplify the tap ECG 

(0.5mV to 5mV peak to peak) to at least the gain of 1000 

(Najeb et al., 2005). Noises are common for tap ECG 

signal from body skin surface. Therefore, this SCC is 

build with instrumentation amplifier (INA121P) with low 

noise differential signal acquisition in order to minimize 

noise interference. The tap ECG is connected to scopemeter 

(fluke, model-199b) at SCC output, as shown in 

Figure 14(a). This shows that, the SCC is operational. 

Then the tap ECG signal from SCC output was capture 

and converted by 8 bit Analog to Digital Converter chip 

set (ADC0820) and translated by PIC18LF452 into GLCD 

code. As a result, the tap ECG signals are able to display 

at the GLCD as shown in Figure 14(b). 


The output from SCC is connected to codec chip set to 

interface with TMS320VC33 in the main board. PCM3003 

is selected because it is fully controlled by hardware 

setting and the data formats are selectable (16 bit or 20 

bit) by physical pins configuration, but in this work the 

data format 16 bit is selected. Then the data transmission 

is only clock in and clock out in sequence from the codec 

chip set. So, it is not dependent on software setting or 

external clocking for operation. 

The connected circuitry depicted in Figure 14 shows 

that the ECG monitoring signal is directly send to 

microcontroller without interrupting the TMS320VC33. 

This means that, the TMS320VC33 is only focus on 

incoming ECG from codec chip set for signal processing. 

Figure 15 depicted two different tap ECG monitoring 

displays result for test subject resting in static and non 

static condition. 

Validation result: Q Onset and T Offset 

The validation of the prototype DSP hardware performance 

for QT algorithm is carried out on the PTB 

diagnostic ECG database by utilizing the annotations 

prepared by Christov et al. (2006). Table 1 shows the


a b 

SCC 

Codec 

ADC 

Analog Front End unit 

Main Board 

Remote socket 

GLCD 


DSP 

RAM 

Signal monitoring 

displays unit 

USB 

ROM 

Signal processing 

with 

Memory unit 

Figure 14. The tap Electrocardiogram capture by scope-meter (fluke, model-199b) and display as monitoring signal 

on Graphic LCD. 

selected 15 files for healthy control subjects with 

annotated Q onset (in millisecond - ms) used in validating 

the prototype DSP hardware. The obtained deviations 

mean errors for prototype DSP hardware is 9.13 ms. 

The prototype design has been tested with algorithm 

for QT analysis and it is validated with 15 healthy 

subject’s files from PTB diagnostic ECG data base. The 

recording length of each selected file by default is about 

one minute trace. Each selected test file data contain one 

selected beat marked visually by referee’s annotation 

(Christov et al., 2006). Table 1 shows that, there are 

selected 15 healthy subject files from the PTB Diagnostic 

ECG database are used to obtain the measured results 

from prototype DSP hardware. The measured results of 

the prototype DSP hardware were presented in line graph 

as depicted in Figure 16. The yellow illustrates the 

annotated Q onset and the purple illustrates the prototype 

DSP hardware. The overall measured results trend in 

Figure 16 shows fluctuation with slight increased towards 

the end of the files listing. However, the comparison of 

the measured results with the annotated Q onset shows 

moderate differences. There are five patient file shows 

higher deviation, which is more than 10 ms and above. 

Those patient files are patient166/s0275lre, patient172/ 

s0304lre, patient180/s0561_re, patient198/s0402lre and 

patient263/s0499_re. 

This is the second part of validation for the prototype 

DSP hardware, which is validate on T offset detection. 

The total of 15 patient files was acquired from PTB 

diagnostic ECG database, T offset detection for each of

(a) Test Subject: Resting in static 

condition 

(b) Test Subject: Resting in non-static condition 

Figure 15: ECG based line moment due to test subject body movement. (a) 

In static condition. (b) In non-static condition. 

Table 1. Comparison of Q onset between the annotated data set with the measured 

one and the obtained deviation results for prototype DSP hardware. 

PTB Diagnostic ECG Data Base Prototype DSP Hardware 

No Patient File 

Q onset (ms) 

Annotated 

Q onset (ms) 

Measured Deviation 

1 patient104/s0306lre 750 756 6 

2 patient116/s0302lre 1136 1140 4 

3 patient165/s0322lre 1221 1224 3 

4 patient166/s0275lre 1632 1612 20 

5 patient172/s0304lre 1470 1486 16 

6 patient180/s0561_re 1751 1732 19 

7 patient198/s0402lre 1165 1192 27 

8 patient233/s0457_re 1543 1540 3 

9 patient234/s0460_re 1518 1520 2 

10 patient241/s0469_re 1717 1716 1 

11 patient241/s0470_re 1566 1568 2 

12 patient243/s0472_re 1155 1152 3 

13 patient248/s0481_re 1061 1056 5 

14 patient263/s0499_re 1473 1496 23 

15 patient264/s0500_re 1389 1392 3 

Mean error 9.13 

Seng et al. 1825


millisecond (ms) 

2000 

1800 

1600 

1400 

1200 

1000 

800 

600 

400 

200 

0 

750 

756 

1136 

1140 

1632 

1612 

1221 

1224 

1470 

1486 

Annotated Data set Prototype DSP Hardware 

1751 

1732 

1165 

1192 

1543 1518 

1540 1520 

1717 1566 

1716 1568 

1155 1061 

1152 1056 

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 

Patient File 

Figure 16. Comparison of annotated (yellow) and the measured result, Q onset in ms (millisecond). Purple illustrates the measured Q 

onset on prototype DSP hardware. 

Table 2. Comparison of T Offset between the annotated data set with the measured one and the 

obtained deviation results for prototype DSP hardware. 

PTB Diagnostic ECG Data Base Prototype DSP Hardware 

No Patient File 

T Offset (ms) 

Annotated 

T Offset (ms) 

Measured Deviation 

1 patient104/s0306lre 1136 1132 4 

2 patient116/s0302lre 1540 1525 15 

3 patient165/s0322lre 1612 1612 0 

4 patient166/s0275lre 2043 2036 7 

5 patient172/s0304lre 1878 1876 2 

6 patient180/s0561_re 2123 2122 1 

7 patient198/s0402lre 1530 1528 2 

8 patient233/s0457_re 1939 1932 7 

9 patient234/s0460_re 1950 1948 2 

10 patient241/s0469_re 2114 2112 2 

11 patient241/s0470_re 1962 1960 2 

12 patient243/s0472_re 1476 1468 8 

13 patient248/s0481_re 1448 1440 8 

14 patient263/s0499_re 1829 1818 11 

15 patient264/s0500_re 1797 1794 3 

Mean error 4.93 

the file was annotated with a marker by utilizing the 

annotations prepared by Christov et al. (2006). Table 2 

depicted the 15 healthy control subjects files with 

annotated T offset (in millisecond - ms) used for 

validating the prototype DSP hardware. The measured T 

offset results for the deviation mean errors are 4.93 ms. 

The measured results of 15 healthy subject files (PTB 

Diagnostic ECG database) from Table 2 for the prototype 

1473 

1496 

1389 

1392 

DSP hardware were presented in line graph as depicted 

in Figure 17. The blue illustrates the annotated T offset 

and the purple illustrates the prototype DSP hardware. As 

an overall trend, it is clear that, there were wild 

fluctuations in T offset measured results towards the end 

of the files listing in Figure 17 which represented the 

prototype DSP hardware. However, the measured results 

of the prototype DSP hardware are close to the

millisecond(ms) 

2500 

2000 

1500 

1000 

500 

0 

1136 

1132 

1540 

1525 

1612 

1612 

2043 

2036 

Annotated Data set Prototype DSP Hardware 

1878 

1876 

1950 

2123 1948 

2122 

1939 

1932 

1530 

1528 

2114 

2112 

1962 

1960 

1476 

1468 

1448 

1440 


1829 

1818 

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 

Patient File 

1797 

1794 

Figure 17. Comparison of annotated (blue) and the measured result, T offset in ms (millisecond). Purple illustrates the 

measured T offset on prototype DSP hardware. 

annotated T offset. It can be clearly seen that, the highest 

deviation of measured results is the second file, 

patient116/s0302lre which is 15 ms, and the lowest or no 

deviation is the third file, patient165/s0322lre which is 0 

ms. 

DISCUSSION 

The prototype DSP hardware has been built, this 

prototype is validated using PTB ECG diagnostic 

database and the acquired tap ECG is able to put on 

Graphic LCD. In summary, the completed prototype DSP 

hardware system is depicted in Figure 18. The Analog 

Front End (AFE) unit is interfaced with volunteer subject 

as shown in Figure 18(b) and Figure 18(d). The output of 

AFE is seen at the oscilloscope screen in Figure 18(e), 

this shows that it is operational. The signal processing 

with memory unit is shown in Figure 18(a). However, in 

Figure 18(c) is the signal monitoring displays unit to 

display the output of AFE on GLCD with various ECG 

based line moment pattern either the volunteer subject is 

resting in static condition or non static condition, the 

display results are shown in Figure 18(f) and Figure 

18(g). The validation process on the prototype DSP 

hardware for QT algorithm is able to be displayed on the 

computer screen as shown in Figure 18(h) through the 

emulator. 

Nevertheless, the ECG data for this prototype DSP 

hardware validation is taken directly from PTB diagnostic 

ECG data base therefore is not influenced by the codec. 

Codec is introduced initially in the system as an initial 

development for this system. The codec will be replace 

with ADC in the future work for real time QT 

measurement on standalone DSP platform. Furthermore, 

the functional test as depicted earlier shows that this 

prototype is operational. Since with the proposed of this 

prototype DSP hardware is targeted at bridging the gap 

among researcher and cardiologist have for in 

collaboration to develop and implementation of QT 

algorithm at a low cost standalone DSP platform. As well 

as eligibility of on board pre-program algorithm for QT 

analysis which existing ECG monitoring equipment are 

lacking. For the presented prototype DSP hardware 

design can be extended in term of design, method and 

structure to capture and conduct other types of 

biomedical signal processing in real time processing. 

Conclusions 

In this paper, we present a design and development of 

standalone DSP prototype for QT Interval processing and 

monitoring. The high cost and lack of flexible applications 

on existing DSP hardware in the market has prompted 

researcher to design and fabricate a low cost with applications 

versatility DSP hardware. By the implementation 

of an operational standalone prototype DSP hardware 

design, the stand alone prototype DSP hardware system 

using floating point digital signal processor has been 

successfully built. The building of the system begins by 

fabricating the main board as the core processing unit 

and the analog front end device as the input. The core 

unit for main board signals processing is designed with a 

Digital Signal Processor (DSPs), TMS320VC33. The QT 

algorithm is applied in this prototype DSP hardware, 

whereby it is validated with PTB diagnostic ECG data 

base and the mean error results are; Q onset is 9.13 ms


h 

d 

Prototype DSP hardware ECG Monitor 

a 

Figure 18: The operational and functional of Prototype DSP hardware 

Figure 18. The operational and functional of Prototype DSP hardware. 

(millisecond) and T offset is 4.93 ms (millisecond). 

REFERENCES 

Badilini F, Erdem T, Zareba W, Moss AJ (2005). ECG Scan: a method 

for conversion of paper electrocardiographic printouts to digital 

electrocardiographic files, 38: 310-318. 

Christov I, Dotsinsky I, Simova I, Prokopova R, Trendafilova E, 

Naydenov S (2006). Dataset of manually measured QT intervals in 

the electrocardiogram, 5: 31 

Christov II S (2006). Fully Automated Method for QT Interval 

Measurement in ECG, 33: 321-324. 

Gorup Ž, Štajer D, Noč M (2000). Signal Analysis Systems for Optimal 

Timing of Electrical Defibrillation, 2: 694-697. 

Kaplan AV, Maim DS, Smith JJ, Feigal DA, Simons M, Jefferys D, 

Fogarty TJ, Kuntz RE, Leon MB (2004). Medical Device 

Development: From Prototype to Regulatory Approval, 109: 3068- 

3072. 

Kara S, Kemaloğlu S, Kirbaş Ş (2006). Low Cost Compact ECG With 

Graphic LCD and Phonocardiogram System Design, 30: 205-209. 

Kligfield P, Hancock WE, Helfenbein D E, Dawson JE, Cook AM, 

Lindauer MJ, Zhou HS, Xue J (2006). Relation of QT Interval 

Measurement to Evolving Automated Algorithm from Different 

Manufacturers of Electrocardiographs, 98: 88-92. 

b 

Analog Front End device 

c 

e 

f 

g 

Laguna P, Thakor NV, Caminal P, Jane R, Yoon H R (1990). New 

algorithm for QT interval analysis in 24 hour Holter ECG: Perform. 

Appl., 28: 67-73. 

Malarvili MB, Hussain S, Ab-Rahman AR (2005). Development of 

Automated 12 Lead QT Dispersion Algorithm for Sudden Cardiac 

Death, 2: 2. 

Malik M (2004). Errors and Misconceptions in ECG Measurement Used 

for the Detectection of Drug Induced QT Interval Prolongation, 37: 

25-33. 

Mneimneh MA, Povinelli, Johnson MT (2006). Integrative Technique for 

the Determination of QT Interval, 33: 329-332. 

Mohammad S, Lim Z, Lin Y, Sani A, Zanjani K, Paracha M, Jenkins J 

(2003). Real time ECG Analysis Using a TI TMSC54X Digital Signal 

Processing Chip, 30: 541-544. 

Moraes J, Viana S (1997).A New Automatic Method to Estimate the QT 

Interval from the Electrographic Signal, 24: 493-496. 

Najeb JM (2007). Real-Time Implementation of Twelve-Lead Automated 

Electrocardiogram System Measurement for QT Dispersion Analysis. 

Univerisiti Teknologi Malaysia: Masters Thesis. 

Najeb JM, Ruhullah A, Salleh Sh H (2005). 12 Channel USB Data 

Acquisition System For QT Dispersion Analysis. pp. 83-86. 

Online available, Physikalisch-Technische Bundesanstalt (PTB) ECG 

data base. URL, http://www.physionet.org/cgi-bin/atm/ATM 

Oweis R, Hijazi L (2006). A Computer aided ECG diagnostic tool, 81: 

279-284. 

Sandham W, Hamilton D, Laguna P, Cohen M (2007). Advances in

Electrocardiogram Signal Processing and Analysis. 2007: 1-5. 

Saviola J (2005). The FDA’s role in medical device clinical studies of 

human subjects. S1-S4. 


Soon IY, Yeo CK, Ng HC (2001). An analogue video interface for 

general purpose DSP. 25: 33-39.



DOI: 10.5897/SRE11.2187 



Effective techniques in drilling to improve the recovery 

process of hydrocarbons in oil and gas sector 

Shafqat Hameed 1 and Muhammad Abbas Choudhary 2 

1 National University of Science and Technology (NUST), Islamabad, Pakistan. 

2 University of Engineering and Technology (UET), Taxila, Pakistan 


The purpose of this research is to use effective method of drilling to improve the recovery process of 

hydrocarbons in oil and gas sector. Hydrocarbons are present in large amount beneath the earth. By 

using exploration geo physics, the reservoir of hydrocarbons below the earth can be located with more 

precision and accuracy. The first step consists of the geological and the seismic survey to obtain the 

image of the beneath rocks. In the next, step drilling is performed to reach the desired depth. 

Depending on the types of rocks beneath the earth the drilling performed may be directional, vertical or 

horizontal. This paper will focus on “how multilateral/horizontal drilling can improve the recovery of low 

permeable hydrocarbons”. Now a day’s offshore drilling is also performed. 

Key words: Hydrocarbons, seismic survey, multilateral drilling, offshore drilling. 

INTRODUCTION 

Hydrogen and carbon forms hydrocarbon which is an 

organic compound in which Crude oil is the most natural 

form of hydrocarbon (Clayden and Greeves, 2001). 

Hydrocarbon produces lot of energy when burnt and are 

the main source of electric energy and home heating. 

Hydrocarbons extracted may be either in liquid or 

gaseous form such as petroleum or natural gas. To 

gather economic values of petroleum, many important 

geological fundamentals and processes are needed 

(Magoon, 1994). Seismic methods are then used to find 

the structure of the layers beneath the earth. Then by 

using information from the structural layers, the point of 

drilling is defined. First oil well was drilled by Col. Edwin 

Drake in USA on August 26, 1859. The well depth was 21 

m only and it took most of the summer to reach the 

desired depth. (Magoon and Beaumont, 1999) 

To recover the hydrocarbons drilling is performed. But 

sometimes the usual drilling methods cannot be applied 

due to certain hurdles present beneath the Earth and 

different location of the reservoirs. The difficulty and cost 

*Corresponding author. E-mail: 

shafqat.hameed@ceme.nust.edu.pk, 

hameed.shafqat@yahoo.com. 

associated becomes very high to reach the desire 

reservoirs by conventional methods. So further drilling 

cannot be performed by using the conventional drilling 

methods. Here technology comes to rescue. The 

effective techniques employ directional, horizontal, and 

multilateral drilling (Figure 1). 

Oil drilling is the most common source of recovery of 

Petroleum. According to Guerriero (2011), “The step 

performed after the structural Geology is reservoir 

characterization (mainly in terms of porosity and 

permeable structures)”. The main sources of Petroleum 

(fossil fuel) are fossilized organic zooplankton and algae 

(Kvenvolden, 2006). They were settled at the bottom of 

sea and lakes mixing with sediments under anoxic 

conditions. 

When adequate thermal energy is passed on to the 

sedimentary organic matter to break chemical bonds, 

petroleum is produced from source rocks. The petroleum 

then starts to migrate along the fault zones. Certain 

numbers of traps are present in the layer of earth where 

petroleum starts to accumulate. They are called reservoir 

rocks. Further movement of petroleum is then prevented. 

These formations are located with the help of geological 

or seismic survey. 

When a certain area of interest has been designated, a 

survey grid will be drawn up with enough detail to allow

Figure 1. Structure of Methane. Source: (Bowling, 1986). 

the interpretation team of geophysicists and geologists to 

map the subsurface strata. A survey ship, complete with 

navigation, recording and shooting equipment (and any 

other surveys which may be run simultaneously such as 

gravity and magnetic), will then follow this grid recording 

the seismic data on computer-compatible tapes for 

further processing. 

Seismic methods use two phenomenon, reflection and 

refraction. They are based on Snell's law. The situation is 

directly analogous to optics, where series of shock waves 

move in the form of wave fronts from the point of energy 

release. When one of these wave fronts encounters an 

interface of two dissimilar materials a process of 

reflection and refraction takes place. From this survey, 

the point of drilling of an oil/gas well is located. 

Vertical drilling 

Vertical drilling is traditional type of drilling in oil and gas 

drilling industry. Drilling a Vertical well is usually cheaper 

than drilling a horizontal well. The production obtained 

from vertical wells is lesser as compared to the horizontal 

wells. The limitation of vertical well is that frequency of 

intersecting a large number of fractures is reduced so 

less production is obtained. 

Horizontal drilling 

Horizontal drilling is the same as vertical drilling until the 

“kickoff point” which is located just above the target oil or 

gas reservoir. From that point, the well is deviated from 

the vertical direction to horizontal. Some limitations of 

horizontal drilling are: 

(a) Horizontal wells have a greater footprint compare to 

multilateral wells. 

(b) When large number of pay zones are present, more 

than one horizontal well are required which is very costly. 

Multilateral drilling 

Hameed and Choudhary 1831 

In this type of drilling, more than one branch emerge from 

a single mother wellbore. The branch may be horizontal, 

vertical or inclined depending upon the availability of the 

zones. More zones can be perforated through multi 

lateral drilling from a single mother well. 

Directional drilling 

It is an angle drilling. Drill pipe provides mechanical and 

hydraulic connection between drilling rig at surface and 

down hole directional steering system. Drill pipe provides 

axial load and disseminate the borehole by destroying the 

rocks. Drill pipe is then pumped with the mud fluid cool 

and lubricate the rock destruction process and to 

transport the rock cutting to the surface (Bowling, 1986). 

Plasma channel drilling: According to Martin (1960), 

“Plasma channel drilling is a process in which sub micro 

second electrical breakdown of rocks is used for efficient 

fragmentation of rock Formations. This preferential 

electrical breakdown of rock is achieved by the use of a 

dielectric liquid (water for example) as the drilling fluid. 

The drilling fluid in the PCD process serves as a superior 

electrical insulator due to differences in the electrical 

properties between the rock and the dielectric liquid 

under impulse conditions. The peak power generated at 

the drill head is typically hundreds of MW. Such high 

powers enable pressures of several GPa to be developed 

in the breakdown channel (Figure 2). (Braun and 

Burnham, 1993). 

METHODOLOGY 

The aim of the study is to check how effective techniques in drilling 

can improve the recovery of hydrocarbons. The research is carried 

out to see whether multilateral drilling can help in the improvement 

or not. In order to carry out the survey, 10 to 15 persons working in 

oil and gas companies are selected including:


1. General Manager drilling operations and services 

2. Manager drilling 

3. Senior drilling superintendent 

4. Geological survey experts. 

Figure 2. Drilling method Source: (Bowling, 1986). 

A qualitative analysis was carried out to determine the cost 

effectiveness of the drilling techniques, the time efficiency and the 

limitations associated with them. It also put light on the difficulties 

and the specific requirements for the drilling techniques which are 

very effective but cannot be employed. The advantages of the latest 

technology over the conventional methods are also highlighted. 

The analysis consisted of qualitative technique. The search will 

include the expert review and cognitive interviews. The result 

obtained from this search will be carried out to draw the final 

conclusions. 

RESULTS 

The purpose of this research was that multilateral drilling 

can improve the recovery of hydrocarbons. Views were 

taken from different persons working in different 

organizations. The results are subsequently. 

Environmental effect 

There are certain wastes associated with the drilling 

process. The waste includes the cuttings of the 

formations and the chemicals used in mud. In order to 

check the effect on environment, 15 persons were 

chosen to take their views. Out of which 10 were of the 

view of thought that multi lateral drilling has less impact 

on the environment, 3 said that it has no effect and 2 said 

that the waste is increased (Figure 3). 

Time effectiveness 

In order to analyze the time associated with the drilling 

process, different persons were chosen. Some were of 

the opinion that time is reduced because single 

foundation is required and dismantling time is also saved. 

However, few said that time taken during the drilling is 

greater because of the complexity. The results are shown 

in Figure 4. 

Cost effectiveness 

In order to check the cost effectiveness, 5 options were 

given to the persons from strongly agree to strongly 

disagree. A rating of 5 was given to highest level and 1 

was given to the lowest. The results are then drawn from

No of views 

the ratings (Figure 5). 

Conclusion 

No. of views 

Figure 3. Effect of waste. 

10 

5 

0 

Time effectiveness 

Figure 4. Time effectiveness. 

Less time greater Greater Time time No effect No effect 

From the views of different persons working in different 

organizations, the following conclusions can be deducted: 

(1) In multilateral wells, higher production is obtained as 

compared to conventional wells. Conventional wells 

(Vertical Wells) yield smaller contact with the reservoir 

when smaller pools are considered, however when 

several branches are laid down from the single mother 

bore, greater contact with the reservoir is achieved. 

Hence high production is obtained. 

(2) Multilateral wells are time and cost effective. For 

multilateral wells, single foundation for the rig is required. 

Conventional wells require separate foundation at every 

place. So the time and cost involved in the dismantling, 

shifting, joining of the rig and the preparation of the 

Hameed and Choudhary 1833 

foundation on the next place is saved. Up to the kick off 

point a single mother well is drilled, it also saves lot of 

cost. 

(3) Environmental impacts are decreased in multilateral 

wells. Different fluids used in drilling process carry out the 

cuttings of drilling. The value of cuttings is reduced in 

multilateral wells so as the drilling fluids. Drilling fluids are 

made up of different chemicals. So when the cuttings and 

fluids are reduced the impact on the environment is also 

reduced. This is a big advantage of multilateral drilling. 

(4) Quick recovery through multilateral wells. Maximum 

number of reservoirs is in contact with multilateral drilling, 

so quick recovery of hydrocarbons is obtained as 

compared to conventional wells. So payback period is 

also smaller. 

(5) Less effect on water underneath the earth surface. In 

multilateral wells, only a single mother bore well is drilled. 

However in conventional drilling method more number of 

wells are drilled. Since mud is used for the cleaning of 

hole continuously. It is made up of different chemicals so


Ratings 

Ratings 

6 

5 

4 

3 

2 

1 

0 

Cost Effectiveness 

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 

Figure 5. Cost effectiveness. 

the effect of multilateral wells on the water beneath the 

earth is reduced. 

(6) It also improves the recovery of low permeable gases 

which cannot be recovered through conventional 

methods. 

LIMITATIONS AND RECOMMENDATIONS 

There are certain limitations associated with multilateral 

drilling. However, certain recommendations will be made 

to enhance the usage of multilateral drilling. 

(1) Multilateral wells are very complex. They are difficult 

to drill. To cope with the complexity, companies must 

employ learning of their workers. Proper training should 

be given to them by hiring experts in that field. 

(2) In multilateral wells, casing is performed only of 

mother bore well, however the branches are open. 

Controlling of the formations is difficult. So drilling should 

be performed only for the developmental wells not for the 

exploratory wells. 

(3) The equipment used for the multilateral drilling is very 

costly. At first glance, this technology looks to be very 

expensive. However, the overall advantage should be 

considered which is much greater when production is 

obtained. 

(4) Well control becomes very difficult when production 

from different branches is obtained. However, by using 

Christmas tree (equipment having separate well heads), 

the problem can be overcome. The production coming 

from branches will go to separate heads. 

Persons 

In summary, the conventional drilling is not much 

complex to perform and that is why it is popular in 

Pakistan. But the production obtained is not in greater 

quantity. On the other hand, multilateral wells are difficult 

to drill and the equipment is expensive but the advantage 

becomes greater when production is obtained. So this 

should be preferred. In Pakistan, no company has 

performed multilateral drilling yet but work is going on for 

the adaptation of this technology. It has bright future in 

Pakistan provided that government of Pakistan 

emphasizes its use and invests heavily in this technology. 

REFERENCES 

Bowling JP, Jogi PN, Burgess TM (1986). “Three Dimensional Bottom 

hole Assembly Model Improves Directional Drilling,” IADC/SPE Conf., 

SPE 14768, Dallas, Feb. 

Braun, Robert L, Burnham, Lan K (1993). “Chemical Reaction Model for 

Oil and Gas Generation from Type I and Type II Kerogen. 

Clayden J, Greeves N (2001) Organic Chemistry Oxford ISBN 

0198503466, p. 21. 

Guerriero V (2011). “Improved statistical multi-scale analysis of 

fractures in carbonate reservoir analogues” 

Kvenvolden KA (2006). “Organic geochemistry – A retrospective of its 

first 70 years”. Organ. Geochem., p. 37. 

Magoon LB, Beaumont EA (1999). Petroleum Systems. In: Handbook of 

Petroleum Geology: Exploring for Oil and Gas Traps, Beaumont, 

E.A.; Foster, H.N. (Eds), American Association of Petroleum 

Geologists Tulsa, pp. 3-1, 3-34. 

Magoon LB, Dow, WG (Eds.) (1994). The Petroleum System-From 

Source to Trap, AAPG Memoir. American Association of Petroleum 

Geologists, Tulsa, 60: 655. 

Martin E (1960). “Experimental investigation of a high energy, high 

pressure or plasma”, J. Appl. Phys., 31: 255-267.

UPCOMING CONFERENCES 

ICPP 2012: International Conference on Plasma Physics Venice, Italy 

November 14-16, 2012 

PRESCO 2012 will be held at the Crowne Plaza Hotel 

Hiroshima, Japan August 6th – 9th, 2012

Conferences and Advert 

July 2012 

5th International Conference on Integrated Systems, Design and Technology (ISDT), 

Kuala Lumpur, Malaysia, 6 Jul 2012 

36th Annual Scientific Meeting of The Human Genetics Society of Australasia 

(HGSA), Canberra, Australia, 22 Jul 2012 

Yeast Genetics Conference, Princeton, USA, 31 Jul 2012 

August 2012 

The Inaugural Pacific Rim Energy & Sustainability Congress, Hiroshima, Japan, 6 

Aug 2012 

November 2012 

4th International Conference on Science in Society, Berkeley, USA, 15 Nov 2012

Scientific Research 

and Essays 

Related Journals Published by Academic Journals 

■ Journal of Oceanography and Marine Science 

■ Journal of Internet and Information Systems 

■ African Journal of Pure and Applied Chemistry 

■ International Journal of the Physical Sciences 

■ Journal of Geology and Mining Research

Download complete issue (pdf 3360kb) - Academic Journals

Create successful ePaper yourself

Delete template?

Save as template?