Recursive state estimation for a class of quantized coupled complex networks subject to missing measurements and amplify-and-forward relay

• Published in: Information sciences Vol. 630; pp. 53 - 73
• Main Authors: Jia, Chaoqing, Hu, Jun, Yi, Xiaojian, Liu, Hongjian, Huang, Jinpeng, Cao, Zhipeng
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.06.2023
• Subjects: Amplify-and-forward relay; Complex networks; Missing measurements; Optimized state estimation; Uniform quantization; Amplify-and-forward relay; Missing measurements; Optimized state estimation; Uniform quantization; Complex networks
• ISSN: 0020-0255; 1872-6291
• DOI: 10.1016/j.ins.2023.02.017

This paper investigates the algorithm design problem of recursive state estimation (RSE) for a class of complex networks (CNs) subject to quantized coupled parameter, missing measurements (MMs) and amplify-and-forward (AF) relay. In the node-to-node network channels, the signals before entering into the communication networks are quantized. In addition, a series of Bernoulli random variables is employed to model the phenomenon of MMs and an AF relay is deployed in the sensor-to-estimator network channels with the purpose of achieving the task of remote data transmission. A recursive state estimator is constructed such that, for all quantized coupled signal, MMs and AF relay, a state estimation error covariance (SEEC) upper bound (SEECUB) is presented and then the estimator gain (EG) is parameterized by optimizing the trace of SEECUB. Subsequently, a rigorous theoretical analysis is given to establish the monotonicity relationship between the trace of the minimized SEECUB and the probabilities of MMs. Finally, a simulation study is carried out for the proposed RSE approach to demonstrate the feasibility and validity of such state estimation strategy. •The recursive state estimation issue is addressed for coupled complex networks under amplify-and-forward relay communication.•New optimized state estimation scheme is given via parameterizing estimator gain in a proper way.•The monotonicity analysis regarding the trace of upper bound and probabilities of missing measurements is presented.