Blind Equalization in Neural Networks

4Research of Blind Equalization Algorithms Based on the FBNN

Abstract: In this chapter, the basic principles and structure types of feedback neural network (FBNN) are introduced. The blind equalization algorithms based on bilinear FBNN, diagonal recursive neural network, and quasi diagonal recursive neural network are studied. The principle characteristics and state functions of these algorithms are analyzed. The iteration formulas of algorithms are deduced and the computer simulations are done. At the same time, the idea of variable step size is introduced in these new algorithms. The mean square error nonlinear function is used as step-size control factor, so the blind equalization algorithm based on the variable step-size diagonal recursive ...