Convergence analysis of discrete-time simplified dual neural network for solving convex quadratic programming problems

The convergence property of discrete-time simplified dual neural network for convex quadratic programming is investigated. By choosing a proper Lyapunov function, a sufficient condition for global convergence is obtained. The convergence rate under the condition is also analyzed, and the exponential convergence property under the condition is proved. Simulation verifies the validity of the theoretical results obtained in this paper.