Weighted least-squares blind deconvolution of non-minimum phase systems

摘要

A new algorithm based on a weighted least-squares technique is proposed that allows online deconvolution of non-minimum phase systems with neither knowledge of the system's impulse response nor source statistics except for source signal moments up to the fourth order. Using computer simulations and computational complexity evaluation, the authors illustrate and compare the performances and the features of the proposed method with those of other techniques found in the literature.