Asian Journal of Information Technology

Year: 2016
Volume: 15
Issue: 11
Page No. 1706 - 1712

Performance Comparison of Adaptive Algorithms with Improved Adaptive Filter Based Algorithm for Speech Signals

Authors : M.A. Raja and B. Aruna Devi

Abstract: Noise cancellation is a technique applied to offset the interference in the signals. Noise Cancellation finds a pile of applications like mobile telephones, hearing aids bluetooth receivers and more. It is difficult to find out any signal, especially speech signal in a noisy environment.An adaptive filter plays a lively role in cancelling the noise in the speech signal. Adaptive filters use several algorithms for scaling down the interferences in the signal. Thus an improved Adaptive filter based Noise cancellation for speech signals approach named Variable Step Size Normalized Differential LMS (VSSNDLMS) Algorithm that incorporates the performance and features of Variable Step Size LMS (VSSLMS) and Normalized Differential LMS (NDLMS) is offered. It is to analyze and compare the performance of the proposed VSSNDLMS Algorithm with various Least Mean Square adaptive algorithms. This algorithm aims in applying the proposed algorithm for various real time applications like auditorium, automobile, seminar hall, etc. The analysis indicates that the proposed adaptive algorithm has fast convergence rate, tracking ability, reduced Mean Square Error (MSE) and maladjustments which is the required characteristics of an adaptive filter. For performance, analysis different input speech signals and sound signal are analyzed. The simulation results show that the proposed VSSNDLMS algorithm converges fast with Minimum MSE and is useful in anticipating the performance of adaptive filters.

How to cite this article:

M.A. Raja and B. Aruna Devi, 2016. Performance Comparison of Adaptive Algorithms with Improved Adaptive Filter Based Algorithm for Speech Signals. Asian Journal of Information Technology, 15: 1706-1712.

Design and power by Medwell Web Development Team. © Medwell Publishing 2024 All Rights Reserved