Robust H∞ filtering for networked control systems with markovian jumps and packet dropouts

Abstract

This paper deals with the H∞ filtering problem for uncertain networked control systems. In the study, network-induced delays, limited communication capacity due to signal quantization and packet dropout are all taken into consideration. The finite distributed delays with probability of occurrence in a random way is introduced in the network.The packet dropout is described by a Bernoulli process. The system is modeled as Markovian jumps system with partially known transition probabilities. A full-order filter is designed to estimate the system state. By linear inequality approach, a sufficient condition is derived for the resulting filtering error system to be mean square stable with a prescribed H∞ performance level. Finally, a numerical example is given to illustrate the effectiveness and efficiency of the proposed design method.