Robust Adaptive Filtering Based on the Hyperbolic Tangent Exponential Kernel M-Estimator Function for Active Noise Cancellation

The standard least-mean square algorithm (LMS) is a simple and efficient approach to apply in noise cancellation problems. However, in the presence of non-Gaussian noises, the standard LMS cannot reach satisfactory performance. For this reason, robust adaptive filtering techniques have been investigated in the last few decades. This paper proposes a robust adaptive filtering approach for noise cancellation. In particular, the developed method is based on the hyperbolic tangent exponential generalized Kernel M-estimate function (HEKM) which achieves a good performance in terms of Average Noise Reduction (ANR). The results demonstrate the efficiency of the proposed algorithm to suppress the presence of the additive spurious signals such as α-Stable noises.