Research On Filtering For Local Strongly Coupled Systems

With the rapid advances in networking and communication technologies, the nodes in complex systems could communicate with each other easily. The local strongly coupled system is such a system that each node traces some targets motion through communications with other nodes in the system, thus the whole system has an intelligence to perform complicated tasks. For simplicity of realization, the exchanged information between subsystems is assumed accurate in study. However, the interactions between nodes are random and there are physical or topological spatial differences.In this paper, we study the filtering method of local strongly coupled systems with packet dropout and time-delay. We introduce the research background and some basic problems of local strongly coupled systems, we discuss the filtering problem of this kind of system especially. Then, we model the local strongly coupled systems with packet dropout, time-delay and random time-delay respectively.By using state augmentation, LMI theory and kalman filtering method, we design robust filter and kalman filter for such kind of systems.The study focus on three aspects: 1) We model the local strongly coupled systems with packet dropout only and study the kalman filtering of the systems, then design the H-inf robust filter for the system; 2) We model the local strongly coupled systems with packet dropout and time-delay, then give the kalman filtering method of the system; 3) We model the local strongly coupled systems with packet dropout and random time-delay ,the design the kalman filter for the systems.