A Resilient Approach to Distributed Recursive Filter Design

The state estimation or filtering problem has proven to be one of the fundamental issues in signal processing and control engineering, and a number of algorithms have been proposed in the literature, see, e.g., [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]. Accordingly, a core problem with the widespread applications of wireless sensor networks (WSNs) is to estimate the plant states based on noisy measurement outputs from distributed nodes. A seemingly natural way is to employ the traditional Kalman filters by establishing a fusion center in WSNs in order to collect all the measurements from the individual sensors and then process the measurements in a global sense. Unfortunately, due to the limited communication capability and energy supply, it might be impossible for the sensors to persistently forward the local messages to the fusion center. As such, the so-called distributed estimation scheme would be more preferable whose main idea is to estimate the plant states based on both the local and the neighboring information according to the topologies of WSNs. Recently, various types of consensus protocols have been proposed with an aim to improve the efficiency of the distributed computation and a rich body of literature has been available on the consensus-based distributed filtering strategies, see, e.g., the seminal work in [11].