Sampled-data exponential stabilization for delayed complex valued neural networks with quantized intermittent communication

This paper addresses the sampled-data exponential stabilization problem for delayed complex-valued neural networks (DCNNs) over capacity-limited communication channels. To reduce communication load, a quantized intermittent sampled-data transmission (QIST) protocol is proposed. A transmission-interval-dependent Lyapunov functional is constructed, which is continuous and piecewise-defined in accordance with the communication mechanism. Using the Lyapunov functional method combined with inequality estimation techniques, two exponential stabilization criteria are derived in the form of complex-valued linear matrix inequalities (LMIs). These criteria establish, for the first time, an explicit quantitative relationship between the duty cycle of the rest intervals and the maximum allowable upper bound of the transmission length. Specifically, it is shown that the maximum allowable duty cycle Ψ ― decreases monotonically as the transmission length increases, and a one-dimensional search algorithm is provided to determine either Ψ ― or the transmission length h k . Numerical simulations on a third-order DCNN demonstrate that the proposed QIST protocol achieves global exponential stabilization with reduced communication resource consumption, and the trade-off between actuator rest time and switching frequency of electronic devices is quantitatively analyzed. These results confirm the effectiveness and practical advantages of the proposed approach.