Blind Equalization Algorithm for Underwater Acoustic Channel Based on Support Vector Regression

In order to overcome the inter-symbol interference generated by multipath effect and improve the frequency band utilization ratio in the underwater acoustic communication system, the blind equalization problem of the underwater acoustic channel is expressed as the support vector regression problem. The error function of constant modulus algorithm (CMA) and multi-modulus algorithm (MMA) are included in the penalty term of the SVR, and the iterative reweighted least squares (IRWLS) method is used to find the optimal equalizer coefficients, thus two batch blind equalization algorithms are proposed in this paper. The simulation experiments show that the proposed two batch algorithms can effectively realize the blind equalization of underwater acoustic channel, compared with the traditional online blind algorithm, they achieves fast convergence with small samples and have excellent blind equalization performance.