Protocol-scheduled nonfragile H∞ filtering for networked T-S fuzzy systems with multipath quantization

Abstract

This paper addresses the challenge of nonfragile $H_{\infty }$ filtering for networked Takagi-Sugeno (T-S) fuzzy systems subject to hybrid network-induced constraints, including limited communication bandwidth and signal quantization. A multipath dynamic quantization strategy is applied to both measurement and estimated outputs to mitigate quantization effects, while a weighted try-once-discard (WTOD) protocol is employed to schedule data transmission and reduce network congestion. By integrating the scheduling protocol and filter design within a unified co-design framework, we formulate the problem as a nonfragile $H_{\infty }$ filtering task under nonideal communication. Using an enhanced Lyapunov-Krasovskii functional approach and average dwell-time (ADT) technique, we derive delay-dependent stability criteria that ensure the resulting filtering error system (FES) is asymptotically stable with a prescribed $H_{\infty }$ disturbance attenuation level. The proposed filter is not only robust against external disturbances but also nonfragile to slight variations in its own implementation gains. Simulations on a tunnel diode circuit model validate the proposed method, demonstrating enhanced estimation accuracy and superior $H_{\infty }$ performance compared to prior approaches.