Predefined-time modified function projective synchronization of memristor-based multidirectional associative memory neural networks with time-varying delay

Abstract

This paper is concerned with the predefined-time modified function projective synchronization problem of memristor-based multidirectional associative memory neural networks (MMAMNNs) with time-varying delay. Firstly, a new predefined-time stability theorem is proposed, which imposes more relaxed and effective conditions on the Lyapunov-Krasovskii function (LKF). Secondly, by designing a new feedback control strategy, sufficient conditions for ensuring the predefined-time modified function projection synchronization between master and slave systems are obtained. In addition, by changing the projection factor, the results of this paper can be flexibly extended to various synchronization types, such as complete synchronization, anti-synchronization, and proportional synchronization. Finally, the correctness of the theory is verified through numerical simulations.