Consensus-Based Distributed State Estimation Over Sensor Networks With Encoding-Decoding Scheme: Accommodating Bandwidth Constraints
Chen Gao, Zidong Wang, Jun Hu, Yang Liu, Xiao He
Abstract
In this article, the consensus-based distributed estimation problem is investigated for linear time-invariant systems over sensor networks, where the sensors are required to estimate the system states in a cooperative manner through communication. A novel encoding-decoding scheme (EDS), which consists of two pairs of an innovation encoder/decoder and an estimation encoder/decoder, is proposed on each sensor to compress the data in order to accommodate the bandwidth-constrained network. An EDS-based consensus estimator is designed whose estimation performance is thoroughly discussed. Specially, a necessary and sufficient condition is established to ensure the convergence of the error dynamics of the state estimates, and then the boundedness issue of the size of the transmitted data is examined. Three optimization algorithms are provided for, respectively, the fastest convergence of the error dynamics, the minimization of the estimator gains, as well as the tradeoff between the convergence rate and the estimation deviation. The effectiveness of the developed distributed estimators is finally illustrated by a series of numerical examples.