Minimum-variance recursive state estimation for complex networks with stochastic switching topologies and random quantization under try-once-discard protocol

This article is concerned with the issue of minimum-variance recursive state estimation (MVRSE) for a class of nonlinear dynamical complex networks (NDCNs) with stochastic switching topologies and random quantization under the try-once-discard (TOD) protocol. Two sequences of Bernoulli distributed random variables with given occurrence probabilities are utilized to characterize the stochastic switching manners of network topologies and the randomly occurring quantized output measurements, where the quantization effects are portrayed by the uniform quantizer. Moreover, the TOD protocol is adopted to arrange the order of the information transmission of network nodes so as to alleviate the communication burden and mitigate the network congestions. The focus of the MVRSE issue is to develop a novel state estimation algorithm such that, for all stochastic switching topologies, random quantization effects and TOD protocol, an optimized upper bound of the estimation error covariance is guaranteed by properly designing the estimator gain. In addition, the theoretical proof is derived, which illustrates that the state estimation error is exponentially mean-square bounded under certain conditions. Meanwhile, we also present the related theoretical analysis, which discusses the impact caused by random quantization. Finally, a numerical experiment is utilized to show the validity of the novel MVRSE approach.