Fuzzy sliding mode control algorithm for networks operating TCP-like congestion control

In this work, the theory of sliding model control is employed to develop a stabilizing AQM (active queue management) in supporting TCP (transport control protocol) with delays. With the use of SM (sliding mode) control theory, we analyze the existence of the SM line and the convergence to SM line based on an AQM controller. Due to the chattering problem inhering in SM law as shown in simulation, we introduce a FSM (fuzzy sliding mode) control algorithm based on the SM theory, which can control the link queue length and aggregate arrival rate to the equilibrium points with little chattering. Since the objectives of both are to control the aggregate rate and queue length to the equilibrium points, we prove later that a social optimization problem can be solved when the control objective is achieved. Implementation issues are discussed and the performance of the new controller is shown via ns-2 simulation to outperform other peer scheme.