Adaptive time delay estimation based on least mean square Sigmoid error with the presence of impulsive noises
Wang Bin · Journal of Dalian Maritime University · 2009
To solve the degradation of the least mean square adaptive time delay estimation(TDE) algorithm with the presence of impulsive noises,a adaptive TDE approach(LMSSETDE) was developed based on the least mean square Sigmoid error (LMSSE) since impulsive noises in the application may usually be modeled as the symmetric α-stable distribution.The error signal was transformed by the Sigmoid function,so the effect of outliers on the estimation result was restrained and the second moment of transformed error signal existed.Under the LMSSE criteria,weight coefficient vectors of the filter which simulating the delay effect of channels,were iterated and the peak location of weight coefficients after the convergence was regarded as the relative time delay between two received signals.Simulation results show that the estimation performance of the LMSSETDE algorithm is excellent under Gaussian and non-Gaussian symmetric α-stable distribution noises.Therefore,the least mean square Sigmoid error is a kind of robust optimal criteria.