Distributed Recursive Filtering for a Class of State-Saturated Nonlinear Systems: A Novel Reputation-Aware Mechanism

This article is concerned with the distributed recursive filtering problem for a class of nonlinear time-varying systems over sensor networks (SNs) under state saturations via a reputation-aware mechanism (RAM). The state saturation is considered to describe the inevitable constraints of physical equipment, while the RAM is introduced to eliminate abnormal data caused by sensor faults or malicious attacks. To address these challenges, a reputation-aware distributed recursive filtering (RADRF) algorithm is developed, where a novel reputation model is constructed to assign credibility scores to neighboring sensors, thereby identifying and rejecting unreliable data. Furthermore, a saturation-dependent recursive filter is designed, through which the upper bound of the covariance of filtering error dynamics (UBCFEDs) is determined by solving a recursive matrix equation. The filter gain is subsequently parameterized by minimizing the trace of the UBCFED so as to guarantee the optimality of estimation performance under nonlinear saturation effects and reputation-based weighting. To validate the feasibility and effectiveness of the developed approach, an illustrative case study concerning the indoor localization of a mobile robot is conducted. Simulation results demonstrate that the proposed RADRF algorithm significantly improves estimation accuracy and robustness compared with schemes that neglect the RAM.