Varible Step Size Multi-Layer Neural Network Blind Equalization Algorithm

In communication systems, ISI(Inter-Symbol Interference) and ICI(Inter-Carrier Interference) seriously affect the quality of communication. Blind equalization techniques can effectively solve this problem. In this paper, the fixed step size multilayer neural network blind equalization algorithm is introduced. The steady-state error and convergence speed of the blind equalization algorithm for fixed step multilayer neural network are limited. To solve this problem, an adjustable step size blind equalization algorithm for multilayer neural network is proposed, it is mainly based on the blind equilibrium algorithm of multilayer neural network, and the convergence speed and convergence accuracy of the system are improved by adding the variable step size algorithm and adjusting the parameters of the network structure, which increases the step size at the beginning to speed up the convergence speed and decreases the step to improve the convergence. Simulation results show that the improved algorithm has faster convergence speed and higher convergence accuracy.