Improved Variable Step Size LMS Algorithm Based on Logarithmic Function

An improved variable step size LMS algorithm based on logarithmic function is proposed. By improving the nonlinear-function relationship between the step size and the error signal, this algorithm still keeps the step size in the convergence range even the processing is affected by large errors and strong interferences. Computer simulation indicates that the proposed algorithm not only has low steady-state error and fast convergence speed, but also overcome the defects of large divergence risk and low anti-interference capability in some algorithms.