Secure Self-Triggered Impulsive Control of Stochastic Complex Networks Under Deception Attacks

Deception attacks, as a prevalent form of network assault, can evade monitoring mechanisms and cause severe losses. In this article, the exponential synchronization problem faced by stochastic complex networks (SCNs) under such attacks is investigated via the introduction of a self-triggered impulsive control (STIC) method. Two novel self-triggered mechanisms (STMs) are developed to regulate impulsive control actions without continuous state monitoring, explicitly accounting for both delay-free and actuator delay scenarios. In the absence of actuator delays, a Lyapunov-based STM is proposed to determine impulsive instants directly from previously sampled states while excluding Zeno behaviors by design. When an actuator delay is present, the resulting asynchrony between the triggering instants and impulsive execution instants is systematically addressed by integrating a comparison system-based approach with graph-theoretic analysis. Sufficient conditions for achieving exponential synchronization are derived in terms of self-triggered parameters, actuator delays, impulsive control gains, network topologies, and attack characteristics. The proposed STMs can predict future triggering instants in advance without the need for continuous or periodic event detection capabilities. Finally, a circuit network example is provided to illustrate the effectiveness and applicability of the proposed theoretical results.