DOWNLOAD MY Ph.D Thesis - UNAM

More documents

Recommendations

Info

Chapter 6OPTIMISATION OF CRAALGORITHMS USINGADAPTIVE CSA6.1 IntroductionContention resolution algorithms (CRA) define the set of rules used to resolvecollisions. They play a vital role in the performance of a multi-access reservationprotocol. This is because the faster they resolve collisions, the lower the access delaywill be and the higher the system throughput will become. Initially, CRAs gained muchinterest in the early 1970s for usage in the transmission of radio packets, and especiallyduring the development of the ALOHANET project [1]. Two major candidates weredefined, ALOHA-based algorithms like ‘exponential backoff and p-persistence’ andsplitting tree algorithm, as reported in Section 1.2.3.2. Since then, much research hasbeen devoted to devising efficient contention resolution mechanisms for multi-accessmedia for Local Area Networks (LANs), Metropolitan Area Networks (MAN), satellitenetworks, radio networks and CATV networks [41].In this chapter, we study in detail performance, optimisation and implementation issuesfor the contention resolution algorithm adopted by the DVB/DAVIC protocol. We payparticular attention to the dynamics and operation of the exponential backoff algorithmand the splitting tree algorithm. Furthermore, special emphasis is paid to the design ofadaptive mechanisms, called Contention Slot Allocators (CSA), which dynamicallyadjust the bandwidth used for contention access, significantly increasing the systemperformance when different bounds are considered.6-1
Chapter 6Optimisation of CRA using adaptive CSAs p6.2 Contention Slot AllocatorAs introduced in [61], [96] and [98], the authors have pointed out that the performanceof a multi-access reservation protocol depends more on the overall framing structureand the capacity assigned to the contention and reservation access modes than thedetails of the CRA adopted. In this section we focus on the performance impact whenthe reservation capacity is dynamically adjusted by the use of a slot allocationmechanism.6.2.1 How many contention-slots per signalling frame?After the INA has scheduled a number of reservation slots (RSs) to carry data packets,any number of contention slots (CSs) may then be allocated. When the load of thenetworks is low, very few CSs are required. On the other hand, since the load is low,there will be unused slots that could be used as CSs. As the offered load increases,depending on the length of the packets, more slots will need to be allocated as CSs.The solution to determine how many CSs is rather simple: allocate all slots that are notbeing used for data as CSs. At low traffic loads, many more CSs will be allocated thanthose required. The surplus of CSs significantly decreases the risk of collision ofreservation requests to a very low level, which in turn reduces the access delay for datapackets.This algorithm is a self-regulating mechanism, since if the number of CSs are too low,requests will not reach the INA and as a result more CSs will be automatically allocated.On the contrary, if the number of CSs is too high, more successful requests will reachthe INA and the number of empty slots that can be used as CSs will decrease to aminimum threshold value, which guarantees that at least a few slots will be reserved forcontention access. Thus, the performance of the network is highly dependant on theminimum number of CSs allocated in each signalling frame. In [61], [96] and [98] theauthors did not consider the minimum number of contention slots that should beallocated in each signalling frame, since this would have led them to a low performanceestimation. In the next section, we present a performance analysis for different valuesof the ‘Minimum number of CSs per signalling frame’ parameter using both CRAs.6-2
Page 6 and 7:
I would like to thank all the peopl
Page 10:
CHAPTER 6: OPTIMISATIO OF CRA ALGOR
Page 13 and 14:
FIGURE 5.17 - SYSTEM THROUGHPUT VS.
Page 15 and 16:
LIST OF TABLESTABLE 2.1 - NEXT GENE
Page 17 and 18:
Chapter 1Introduction pBecause of h
Page 19 and 20:
Chapter 1Introduction pA first step
Page 21 and 22:
Chapter 1Introduction passumption i
Page 23 and 24:
Chapter 1Introduction ppolynomial b
Page 25 and 26:
Chapter 1Introduction pIn general,
Page 27 and 28:
Chapter 1Introduction pminislots to
Page 29 and 30:
Chapter 1Introduction pThe authors
Page 31 and 32:
Chapter 1Introduction prate access
Page 33 and 34:
Chapter 1Introduction p1.4 Overview
Page 35 and 36:
Chapter 2OVERVIEW OF CURRENT CATVNE
Page 37 and 38:
Chapter 2Overview of current CATV n
Page 39 and 40:
Page 41 and 42:
Page 43 and 44:
Page 45 and 46:
Page 47 and 48:
Page 49 and 50:
Page 51 and 52:
Page 53 and 54:
Page 55 and 56:
Chapter 3THE DVB/DAVIC PROTOCOL3.1
Page 57 and 58:
Chapter 3The DVB/DAVIC protocol pEu
Page 59 and 60:
Chapter 3The DVB/DAVIC protocol pMA
Page 61 and 62:
Chapter 3The DVB/DAVIC protocol pTh
Page 63 and 64:
Page 65 and 66:
Page 67 and 68:
Chapter 3The DVB/DAVIC protocol p3.
Page 69 and 70:
Chapter 4Simulation and analytical
Page 71 and 72:
Page 73 and 74:
Page 75 and 76:
Page 77 and 78:
Page 79 and 80:
Page 81 and 82:
Page 83 and 84:
Page 85 and 86:
Page 87 and 88:
Page 89 and 90:
Page 91 and 92: Chapter 4Simulation and analytical
Page 105 and 106: Chapter 5Upstream channel capacity
Page 141: Chapter 5Upstream channel capacity
Page 145 and 146: Chapter 6Optimisation of CRA using
Page 171 and 172: Chapter 7PERFORMANCE OPTIMISATION F
Page 173 and 174: Chapter 7Performance optimisation f
Page 193 and 194:
Chapter 7Performance optimisation f
Page 195 and 196:
Page 197 and 198:
Page 199 and 200:
Page 201 and 202:
aChapter 7Performance optimisation
Page 203 and 204:
Page 205 and 206:
Page 207 and 208:
Page 209 and 210:
Chapter 8Final conclusions p8.2 Gen
Page 211 and 212:
Chapter 8Final conclusions pWe have
Page 213 and 214:
Chapter 8Final conclusions pperform
Page 215 and 216:
Chapter 8Final conclusions pslots i
Page 217 and 218:
Chapter 8Final conclusions pfocused
Page 219 and 220:
Chapter 8Final conclusions preserva
Page 221 and 222:
Chapter 8Final conclusions pbandwid
Page 223 and 224:
Chapter 8Final conclusions pThe opt
Page 225 and 226:
Chapter 8Final conclusions p8.4 Fut
Page 227 and 228:
Chapter 8Final conclusions pApplica
Page 229 and 230:
Chapter 8Final conclusions pTCIS st
Page 231 and 232:
Referencesp[17] I. Cidon, and M. Si
Page 233 and 234:
Referencesp[50] N. F. Huang, C. A.
Page 235 and 236:
Referencesp[85] V. Rangel, R. Edwar
Page 237 and 238:
Referencesp[112] K. Sriram, “Meth
Page 239 and 240:
APPEDIX A: GLOSSARYAALABRADSLATMBCB
Page 241 and 242:
Appendix AGlossary pkbpsLANLLCMACMA
Page 243 and 244:
Appendix AGlossary pUSBUWVDSLVoDVoI
Page 245 and 246:
Appendix BProtocol stack and packet
Page 247 and 248:
Appendix BProtocol stack and packet
Page 249 and 250:
APPENDIX C: D VERIFICATION TEST FOR
Page 251 and 252:
Appendix D Verification test p209 2
Page 253 and 254:
Appendix EResults for a 6 Mbps upst
Page 255:
Appendix FGuide to CD-ROM p IEEE802
show all

DOWNLOAD MY Ph.D Thesis - UNAM

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?