A support vector machine blind equalization algorithm based on orthogonal wavelet transform

For greatly overcoming the disadvantages of low convergence rate and local minima of constant modulus algorithm(CMA), a support vector machine blind equalization algorithm based on orthogonal wavelet transformation(WT-SVM) was proposed. The proposed WT-SVM algorithm employs the support vector machine for overcoming the local minima in convergence process and uses orthogonal wavelet transform to reduce the autocorrelation of the equalizer input signals and its tracking capability is improved via selecting a suitable kernel function. Computer simulation with underwater acoustic channels indicates that the proposed algorithm has fast convergence rate and low mean square error and can greatly overcomes the shortcomings of the traditional CMA algorithm.