Sliding Mode Learning Control for Congestion Control of Wireless Sensor Networks

More attention has been paid to congestion control in the emerging area of wireless sensor networks (WSNs). However, most of the existing congestion control schemes are based on the best-effort data transmission service model. When networks congestion occurs, different categories of traffic are discarded indiscriminately, therefore, users' Quality of Service (QoS) is not able to be guaranteed. In this paper, by introducing Differentiated Services (Diffserv) scheme into WSNs, the network is modelled as three classes of traffic services by a validated nonlinear fluid flow model to meet different QoS requirements. The work presented here applies a robust sliding mode learning control (SMLC) algorism to congestion control in WSNs. The simulation results demonstrate that proposed SMLC mechanism mitigates congestion problem effectively with a strong robustness against time-varying delays and has a better transient performance than the conventional sliding mode control (SMC).