Research of Self-Adaptive of TCP Vegas Base on Dynamic Time Delay

Autor: Yueqiu JIANG, Chunqiang YANG, Qixue GUAN
Jazyk: angličtina
Rok vydání: 2013
Předmět:
Zdroj: Sensors & Transducers, Vol 157, Iss 10, Pp 355-362 (2013)
ISSN: 1726-5479
2306-8515
Popis: TCP Vegas is a traditional congestion control algorithm based on measuring round-trip time. The performance of this algorithm is superior to the traditional TCP protocol in many respects. TCP Vegas congestion avoidance mechanisms also have some problems through the study, including competition with Reno whose aggressive performance lower (including the lower aggressive performance in competition with Reno). In order to solve these problems, this paper proposes an improved TCP Vegas congestion avoidance algorithm, TCP Vegas-E. This algorithm dynamically adjusts the alpha and beta thresholds through different network delays to approach the largest number of packet, and to make full use of the available bandwidth on the Internet. The simulation shows that the new congestion avoidance algorithm can get good performance in wired networks to solve the problem that the Vegas - E algorithm is able to enhance the bandwidth competitive ability with Reno in coexistence network environment by the NS-2 simulation software.
Databáze: OpenAIRE