Encoding-decoding-based finite-horizon recursive secure state estimation for dynamic coupled networks with random coupling strength☆

Abstract

The recursive secure state estimator design issue is investigated in this article for a class of dynamic coupled networks (DCNs) through the encoding-decoding scheme. The random variables distributed over specified intervals are utilized to depict the stochastically varying coupling strengths. For the sake of facilitating the data transmission, saving network resources and cutting down security risks, the two-description coding scheme (TDCS) is employed in the communication channels. Based on the TDCS, the encoders are installed at two independent communication channels, and the packet loss phenomena are considered in two encoder-to-decoder channels. The random denial of service (DoS) attack is taken into account when the decoded data are transmitted to the estimator through the shared communication networks. By recursively dealing with two coupled backward Riccati difference equations (CBRDEs), the desired finite-horizon H∞ state estimator gain matrix (EGM) can be determined. Finally, a numerical example is exhibited to illustrate the efficiency of the designed finite-horizon secure state estimator.