Robust $H_{\infty }$ Filtering for Discrete-Time Heterogeneous Networks via Event-Triggered Pinning Control.

Robust $H_{\infty }$ filtering issue for a class of nonlinear discrete-time network systems (NSs) with interval time-varying delay and multiple parameter uncertainties is studied. The model is composed of heterogeneous nonlinear nodes. Moreover, the uncertainties in node dynamics, external disturbances, couplings, and outputs, which cannot be dealt with in one way, are all considered. Based on the uncertain output, asynchronous pinning event-triggered control (ETC) is designed. Since classical Kalman filtering method is not applicable, novel techniques are developed to overcome the difficulties induced by the nonlinear functions, time delays, as well as the multiple uncertainties to derive exponential $H_{\infty }$ stabilization conditions. It is interesting that all the parameters of the filter can be designed though the nonlinear nodes are heterogeneous and only partial nodes are controlled. The merits of the theoretical analysis are demonstrated by numerical simulations.