Skip to Main Content
In many physical communication systems, the channel noise is known through experimental measurements to be non-Gaussian with an essential impulsive phenomena. The performance of standard signal estimators degrades substantially in the presence of such impulsive ambient noise. New robust M-estimation techniques are developed for combating impulsive noise and interference in communication systems. Power functions of the proposed robust estimators are obtained by minimizing a nonquadratic residual function derived from the P.J. Huber's minimax robust estimation theory (see "Robust statistics", Wiley, 1981). Maximum peaks of the power functions are used for estimation of communication signals as well as direction-of-arrival (DOA). Simulation shows that the proposed robust techniques, the robust median and robust Huber estimators, offer significant performance gain over the conventional estimator, with the best results given by the robust Huber estimator.
Date of Conference: 15-18 June 2003