{"title":"Design and analysis of a modified TCP for Optical Burst Switched networks","authors":"Subhasis Datta, Avijan Dutta, S. Choudhury","doi":"10.1109/ICBIM.2014.6970919","DOIUrl":null,"url":null,"abstract":"Performance of existing TCP implementation are not satisfactory in Optical Burst Switched (OBS) networks. TCP throughput suffers in OBS networks due to undesired timer expiry as a result of burst loss in bufferless OBS networks. TCP misinterprets any loss as a signal for congestion and burst assembler introduces different delays to different TCP segments. In this paper we propose a loss differentiation scheme that includes the assembly time in the decision making process. We present a protocol level technique for estimating assembly time at the TCP end points and use this information to differentiate congestion induced loss or contention induced loss.","PeriodicalId":6549,"journal":{"name":"2014 2nd International Conference on Business and Information Management (ICBIM)","volume":"8 1","pages":"7-10"},"PeriodicalIF":0.0000,"publicationDate":"2014-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2014 2nd International Conference on Business and Information Management (ICBIM)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICBIM.2014.6970919","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
Performance of existing TCP implementation are not satisfactory in Optical Burst Switched (OBS) networks. TCP throughput suffers in OBS networks due to undesired timer expiry as a result of burst loss in bufferless OBS networks. TCP misinterprets any loss as a signal for congestion and burst assembler introduces different delays to different TCP segments. In this paper we propose a loss differentiation scheme that includes the assembly time in the decision making process. We present a protocol level technique for estimating assembly time at the TCP end points and use this information to differentiate congestion induced loss or contention induced loss.