Unified Single-Layer Inverse-Free Neurodynamic Network for Solving Absolute Value Equations

This technical note presents a unified single-layer inverse-free neurodynamic network (SINN) to tackle absolute value equations (AVEs). A new global error bound for AVE is established and it is tighter than existing error bounds. Moreover, under different parameter conditions, the global asymptotic, finite-time and fixed-time convergence of the proposed SINN are proved with the help of this new error bound, respectively. Unlike some existing neurodynamic networks, SINN’s distinct feature is simple structure, fixed-time convergence, and smaller upper bound of convergence time. Finally, simulation results are provided to verify the effectiveness and advantages of the proposed approaches.