Addressing performance degradations in end-to-end congestion control has beenone of the most active research areas in the last decade. Active queuemanagement (AQM) aims to improve the overall network throughput, whileproviding lower delay and reduce packet loss and improving network. The basicidea is to actively trigger packet dropping (or marking provided by explicitcongestion notification (ECN)) before buffer overflow. Radial bias function(RBF)-based AQM controller is proposed in this paper. RBF controller issuitable as an AQM scheme to control congestion in TCP communication networkssince it is nonlinear. Particle swarm optimization (PSO) algorithm is alsoemployed to derive RBF parameters such that the integrated-absolute error (IAE)is minimized. Furthermore, in order to improve the robustness of RBFcontroller, an error-integral term is added to RBF equation. The results of thecomparison with Drop Tail, adaptive random early detection (ARED), randomexponential marking (REM), and proportional-integral (PI) controllers arepresented. Integral-RBF has better performance not only in comparison with RBFbut also with ARED, REM and PI controllers in the case of link utilizationwhile packet loss rate is small.
展开▼
