Performance of the Frequency-Response-Shaped LMS Algorithm in Impulsive Noise

The performance of the frequency-response-shaped least mean square (FRS-LMS) adaptive algorithm in estimating a sinusoidal signal in impulsive and correlated noise is investigated. The algorithm does not require a priori knowledge about the nominal Gaussian process and is able to adapt to changes in the environment. The performance of the FRS-LMS is compared to that of the leaky-LMS algorithms in terms of mean square error (MSE) and convergence speed. The results indicate that while the FRS-LMS and the leaky LMS algorithms perform similarly in AWGN, the FRS-LMS provides superior performance in impulsive and correlated noise environments. The performance gain is due to the frequency shaping and outlier reduction properties of the algorithm.