A study of LMS Newton equalization algorithm in nonlinear channel

In this paper,we use Hammerstein model and Wiener model in non-linear structure instead of the Volterra series model in order to reduce the computational complexity,and propose a nonlinear transmission system model constructed by Hammerstein model and Wiener model.Based on the proposed system model,Nonlinear Channel LMS(NCLMS) algorithm,improved algorithm for type 1 NCLMS Newton and improved algorithm for type 2 NCLMS Newton are proposed and derived.The simulation results show that NCLMS algorithm has slow convergence and instable performance,improved algorithm for type 1 NCLMS Newton overcomes the NCLMS algorithm's shortcomings of slow convergence and improves the stability of the algorithm,improved algorithm for type 2 NCLMS Newton inherits the virtues of improved algorithm for type 1 NCLMS Newton and further improve the convergence speed,which is close to ideal NCLMS Newton algorithm.