Extended Mutual Information Separation Algorithms Based on Multi-Hidden Layer
Cai Bang-gui · Telecommunication Engineering · 2013
Extended mutual information separation(EMISEP) algorithm uses a single hidden layer neural network to approximate nonlinear function of cost function,so the adjustable parameter is limited and it needs more iteration times to converge,which leads to relatively slow convergence speed.To overcome this problem,this paper uses double hidden layer perceptions to approximate nonlinear function of cost function,and uses mutual information minimum of separation signals as cost function,which is optimized by gradient descent method.This increases the number of adjustable parameters.The simulation results prove that the improved algorithm has faster convergence speed and smaller error comparing with the original algorithm.