Random early detection gateways for congestion avoidance
IEEE/ACM Transactions on Networking (TON)
Fluid-based analysis of a network of AQM routers supporting TCP flows with an application to RED
Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication
Understanding TCP Vegas: a duality model
Journal of the ACM (JACM)
Feedback Control of Dynamic Systems
Feedback Control of Dynamic Systems
Hi-index | 0.00 |
The growth in the TCP/IP traffic and the importance of the industrial applications using Internet (Flexibible Control and Instrumentation, Telecontrol, Telemedecine,...) as communication medium have made necessary to know better the dynamic behavior of this dynamic system (TCP/IP flow).The objectif is to reduce the congestion and thus render better service to the user with constant investments. Different models and different techniques of congestion control are proposed. These techniques have in common the necessity to estimate the congestion level of a link and to return this information to the transmission sources. The results obtained has proved that to reduce the congestion it is necessary to anticipate it by slowing down the sources before it occurs. In our paper we present a robust congestion control algorithm based on Smith Predictor (SP). The advantage of using SP is its capacity to predict the state of a dynamic system (congestion in our case) and thus reducing the effect of input delay inherent of TCP/IP flows.