Analysis of LMS and NLMS noise cancellations of speech signal using Matlab Simulink / Roziahtushahila Hashim

The LMS adaptive filter has several parameters which can affect their performance. From among these parameters, most papers handle the step size parameter for controlling the performance. This paper presents a study of LMS and NLMS adaptive filter for noise cancellation on speech signal. This analysis is based on the results from simulation of MATLAB Simulink and their outputs from both models are compared in order to investigate the suitable type of adaptive filter algorithm. It is found that the NLMS is a variant of the LMS algorithm which is more robust when implemented in finite-precision hardware and well understood convergence behaviour compare to the other adaptive algorithm.