Robust H-infinity control for uncertain time-delay TCP/AQM network system

Abstract

A robust active queue management (AQM) scheme based on H-infinty theory is presented for the problem of congestion control in TCP communication networks. In TCP/IP networks, the packet-dropping probability function is considered as a control input, and network parameters are time-varying. Thus a TCP/AQM controller is modeled as a uncertain time-delay system with a saturated input. The corresponding existence condition of the observer and controller are obtained by applying Lyapunov-Krasovskii function approach and the linear matrix inequality technique. Simulation results show that the proposed scheme can track the desired queue length very quickly under various network conditions, and avoid the problem of dynamic network congestion.