New Strategy for Congestion Control based on Dynamic Adjustment of Congestion Window
This paper presents a new mechanism for the end-to-end congestion control, called EnewReno. The proposed mechanism is based on the enhancement of both the congestion avoidance and the fast recovery algorithms of the TCP NewReno so as to improve its performance. The basic idea of the proposed mechanism is to adjust the congestion window of the TCP sender dynamically based on the level of congestion in the network so as to allow transferring more packets to the destination. The performance of the proposed mechanism is evaluated and compared with the most recent mechanisms by simulation studies using the well known Network Simulator NS-2 and the realistic topology generator GT-ITM.
Keywords: TCP, Congestion Control, Congestion Avoidance, Fast Recovery algorithm, Network Protocols.
Download Full-Text
ABOUT THE AUTHOR
Gamal Attiya
Gamal Attiya graduated in 1993 and obtained his MSc degree in computer science and engineering from the Menufiya University, Egypt, in 1999. He received PhD degree in computer engineering from the University of Marne-La-Vallée, Paris-France, in 2004. His main research interests include distributed computing, task allocation and scheduling, computer networks and protocols, congestion control, QoS, and multimedia networking.
Gamal Attiya
Gamal Attiya graduated in 1993 and obtained his MSc degree in computer science and engineering from the Menufiya University, Egypt, in 1999. He received PhD degree in computer engineering from the University of Marne-La-Vallée, Paris-France, in 2004. His main research interests include distributed computing, task allocation and scheduling, computer networks and protocols, congestion control, QoS, and multimedia networking.
