Event-triggered filtering for complex networks subject to random coupling strength and missing measurements: A partial-nodes accessible case

In this paper, the event-triggered filtering issue is addressed for a class of discrete-time nonlinear complex networks (CNs) considering random coupling strength (RCS) and missing measurements (MMs) under the condition that partial nodes information is accessible. To begin with, a uniformly distributed random variable over a settled interval is employed to model random changes in coupling strength. Then, a Bernoulli distributed random variable is considered to depict the phenomenon of MMs with uncertain occurrence probability. The event-triggered communication (ETC) protocol is applied to schedule data to relieve transmission burden. Subsequently, a novel partial-node-based recursive filtering algorithm is constructed by taking the influence of RCS, MMs and ETC mechanism into account, in which the filter gain is parameterized to achieve the minimization for the upper bound of filtering error covariance (UBFEC). Besides, performance discussion for the developed filter is provided through strict theoretical derivations, including the uniform boundedness and monotonicity relationship. Finally, two simulation examples are presented to demonstrate the effectiveness of the proposed estimation scheme.