Robust M-estimation techniques for non-Gaussian CDMA wireless channels with phased array antenna

A new robust M-estimation technique is developed for combating impulsive noise and multiuser interference in code-division multiple-access (CDMA) communication channels with a phased array receiver. A special power function of the proposed robust estimator is derived from the Huber's minimax robust estimation theory by minimizing a non-quadratic loss function of estimation residuals. Maximum peaks of this power function are used for estimation of communication signals as well as the direction-of-arrival (DOA). The produced theoretical analysis of the asymptotic accuracy of estimation is used for selection of the optimal minimax loss function. A strong advantage of the developed robust algorithm is a decreased sensitivity of the estimates with respect to actually unknown distributions of random noise and interferences. Simulation demonstrates that the proposed robust algorithms offer significant performance gain over the conventional least squares (LS) based estimator in terms of both BER and maximum user capacity.