A Novel ZNN-Based Chaos Synchronization Controller and Its Application in Secure Voice Communications
Jiajie Luo, Jiguang Li, Lin Xiao, Jichun Li
Abstract
Current variable-convergence-parameter zeroing neural networks (ZNNs), including the VCP-ZNN and the FCP-ZNN, are either inefficient or unintelligent. Although researchers have discussed the application of ZNN in chaos synchronization, these ZNN-based chaos synchronization controllers are rarely used in real-world applications. To the best of the authors’ knowledge, no researchers have applied the ZNN-based chaos synchronization controllers in secure voice communication. In this study, we established a novel chaos synchronization controller based on the proportional–integral-convergence-parameter ZNN (PICP-ZNN) model, which is both computationally efficient and intelligent. It was then used in secure voice communication. To demonstrate the superior features of the proposed PICP-ZNN model, we presented both theoretical analysis and numerical experiments to show its fixed-time convergence, robustness, and adaptiveness. In addition, a detailed comparison with other state-of-the-art variable-convergence-parameter ZNNs was presented to highlight our contribution further. The upper bound of the settling time is also estimated in both noisy and noise-free environments. Overall, this study offers a novel ZNN-based secure communication scheme. The PICP-ZNN models may serve as a novel source of inspiration for enhancing the variable-convergence-parameter ZNN even further.