State Estimation for Stochastic Systems with Coding-Decoding Scheme and Packet Dropouts

In this paper, the state estimation problem is investigated for discrete-time stochastic systems with packet dropouts via a coding-decoding scheme (CDS). It is assumed that only the sequence of limited codewords can be transmitted through digital channels. The random variables satisfying Bernoulli distribution are used to describe the circumstance of packet loss during the transmission of codewords. An improved uniform quantization-based coding-decoding method is proposed, and sufficient conditions which involve quantization level, coding period and packet dropout probability are obtained for the detectability. In addition, the desired observer parameter is derived by solving some linear matrix inequalities. Finally, an example is presented to certify the effectiveness of the designed procedures.