Loss Inference in Unicast Network Tomography Based on TCP Traffic Monitoring

دانشگاه آزاد اسلامی اصفهان(خوراسگان)

منو

رکورد قبلی

رکورد بعدی

" Loss Inference in Unicast Network Tomography Based on TCP Traffic Monitoring "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	151189
Doc. No	:	ET22981
Main Entry	:	Yau-Yau Yolanda Tsang
Title Proper	:	Loss Inference in Unicast Network Tomography Based on TCP Traffic Monitoring
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	Network tomography is a promising technique for characterizing the internal behav-ior of large-scale networks based solely on end-to-end measurements. Despite theefficiency of active probing in most network loss tomography methods, these mea-surements impose an additional burden on the network in terms of bandwidth andnetwork resources. They can therefore cause the estimated performance parame-ters to differ substantially from losses suffered by existing TCP traffic flows. In thisthesis, we propose a promising passive measurement framework based on the sarn-pling of existing TCP flows. We demonstrate its performance using extensive ns-2simulations. We observe accurate estimates of link losses (with 2 mean absoluteerror). We also describe the Expectation-Maximization (EM) algorithm in solvingthe Maximum Likelihood (ML) Estimates-....,....-...,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL46210.pdf
Title and statement of responsibility and	:	Loss Inference in Unicast Network Tomography Based on TCP Traffic Monitoring [Thesis]