The Enhanced ECN Mechanism to Accelerate the Congestion Control Convergence in Network
Weirong Liu, Min Wu, Jun Peng, Guojun Wang · 2009
To solve the network congestion problem, F.P. Kelly proposed proportional fair and classic Kelly model which can converge to proportional fair point with asymptotic stability. Based on it, many new flow control mechanisms were derived to improve its performance. However, Most of their convergence rates are linear since in classic Kelly model, link price is only positive. This paper presents an enhanced ECN mechanism to accelerate the model convergence rate. In which, the link price can be negative and the price can be conveyed by using a new ECN bit without adding extra packet or packet header. NS2 simulation results show that the enhanced ECN mechanism can keep fairness and asymptotic stability with more rapid convergence rate.