Exponential Convergence Flow Control Model for Congestion Control

Recently, many new flow control mechanisms derived from the classic Kelly model are proposed to solve network congestion problems. They perform well in stability, fairness or robustness. However, most of them have convergence rates that are linear since in classic Kelly model, the link price is only positive. In addition, some needs to introduce extra packet headers to get the price information. In this paper, we present a novel flow control model based on Kelly model in which link price can be negative to improve the convergence rate. Further, the proposed model uses two bits of ECN field in IP header to feedback price instead of introducing new packet header data. By this approach, we can implement flow control schemes achieving exponential convergence in traditional TCP/IP datagram format. NS2 simulation results show that our model can keep the advantages of other flow mechanisms, such as fairness and asymptotic stability with more rapid convergence rate