Recent advances on distributed filtering for stochastic systems over sensor networks

Abstract

Sensor networks comprising of tiny, power-constrained nodes with sensing, computation, and wireless communication capabilities are gaining popularity due to their potential application in a wide variety of environments like monitoring of environmental attributes and various military and civilian applications. Considering the limited power and communication resources of the sensor nodes, the strategy of the distributed information processing is widely exploited. Therefore, it would be interesting to examine how the topology, network-induced phenomena, and power constraints influence the distributed filtering performance and to obtain some suitable schemes in order to solve the addressed distributed filter design problem. In this paper, we aim to survey some recent advances on the distributed filtering and distributed state estimation problems over the sensor networks with various performance requirements and/or randomly occurring network-induced phenomena. First, some practical filter structures are addressed in detail. Then, the developments of the distributed Kalman filtering, distributed state estimation based on the stability or mean-square error analysis, and distributed filtering are systematically reviewed. In addition, latest results on the distributed filtering or state estimation over sensor networks are discussed in great detail and some challenges are highlighted. Finally, some concluding remarks are given and some possible future research directions are pointed out.