DSP Simulation of a New Variable Step-size Blind Equalization Algorithm Based on Wavelet Transformation

摘要

In order to overcome the slow convergence rate and large Mean Square Error (MSE) of Censtant Modulus Algorithm (CMA), an equalizer and its output are expressed by a family of orthogonal wavelet packet functions. Then, a new variable step-size blind equalization algorithm based on Wavelet Transformation and Hyperbolic Tangent Function (WTTanh-CMA) was proposed. The convergence rate of the proposed algorithm can be improved via using the de-correlation ability of orthogonal wavelet packet transformation. The optimal parameter values of the proposed algorithm are first obtained by MATLAB simulation. Then the efficiency of proposed algorithm is proved by DSP simulation in underwater acoustic channel.