Particle filter with Markovian packet dropout and time delay

摘要

This technical note is concerned with particle filter for the discrete-time nonlinear networked control system. First, modified particle filter algorithm with Markovian packet dropout and time delay is proposed, and its error covariance is benchmarked by Markovian Cramer-Rao lower bound. Second, an upper bound of the Markovian Cramer-Rao lower bound is presented for some special nonlinear networked systems. Third, some necessary conditions for the boundness of error covariance are given by obtaining some sufficient conditions for the bounded Markovian Cramer-Rao lower bound. Finally, numerical examples are presented to illustrate the efficiency of proposed particle filter. (C) 2018 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.