Noise-suppressing zeroing neural network for online solving time-varying matrix square roots problems: A control-theoretic approach

作者：

Highlights：

• Noise-tolerant neural networks are proposed for time-varying matrix square roots.

• The superiorities are demonstrated for noise-tolerant zeroing neural networks.

• Different activation functions may accelerate the convergence speed.

• The MATLAB Simulink modeling is directly beneficial to the hardware implementation.

摘要

•Noise-tolerant neural networks are proposed for time-varying matrix square roots.•The superiorities are demonstrated for noise-tolerant zeroing neural networks.•Different activation functions may accelerate the convergence speed.•The MATLAB Simulink modeling is directly beneficial to the hardware implementation.

论文关键词：Noise-suppressing zeroing neural network model,Time-varying matrix,Square roots problem,Exponential convergence,Global convergence

论文评审过程：Received 8 January 2020, Revised 24 September 2021, Accepted 20 November 2021, Available online 3 December 2021, Version of Record 4 January 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116272