Adaptive Filtering using Steepest Descent and LMS Algorithm

Pratik Nawani; Akash Sawant; Shivakumar

CALL FOR PAPERS : Mar-2023

Submission Last Date

25-Mar-23

Submit Manuscript Online

FOR AUTHORS

FOR REVIEWERS

ARCHIEVES

DOWNLOADS

Open Access

Adaptive Filtering using Steepest Descent and LMS Algorithm

Author(s):

Pratik Nawani , Mukesh Patel School of Technology Management and Engineering, NMIMS University; Akash Sawant, Mukesh Patel School of Technology Management and Engineering, NMIMS University; Shivakumar, Mukesh Patel School of Technology Management and Engineering, NMIMS University

Keywords:

Steepest Descent, LMS, Mean Square Error, Tap Weights, Stochastic Gradient Algorithm

Abstract:

In many practical scenarios, it is observed that we are required to filter a signal whose exact frequency response is not known. A solution to such problem is an adaptive filter. It can automatically acclimatize for changing system requirements and can be modelled to perform specific filtering and decision-making tasks. This paper primarily focusses on the implementation of the two most widely used algorithms for noise cancelling which form the crux of adaptive filtering. The empirical explanation of steepest descent method is elucidated along with its simulation in MATLAB by taking a noise added signal and applying the ingenuity of this algorithm to get the desired noise-free response. Furthermore, this paper also sheds light on a more sophisticated algorithm which is based on the underlying criteria of minimum mean square error called as the Least Mean Square (LMS). Additionally, there are various applications of adaptive filtering including system identification which is briefly explained to emphasize the instances where it can be used.