Convergence of accelerated modulus-based matrix splitting iteration methods for linear complementarity problem with an [formula omitted]-matrix

Abstract

The theoretical analysis of the accelerated modulus-based matrix splitting iteration methods for the solution of the large sparse linear complementarity problem is further studied. The convergence conditions are presented by fully utilizing the H+-matrix property of the system matrix, and the optimal iteration parameters in accelerated modulus-based accelerated overrelaxation method are determined by minimizing the spectral radius of the iteration matrix. Numerical experiments further confirm the theoretical discussion, and show that the proposed methods accelerate the convergence of the modulus-based matrix splitting methods with less iteration steps and CPU time.