Denoising of Sounds of Musical Instruments by RLS Adaptive Algorithm

Raghavendra Sharma, V. Prem Pyara


Musical instrument sound information in the form of digital signal is becoming very popular in modern communication era, but the signal obtained after transmission is often corrupted with noise. The received sound signal needs processing before it can be used for applications. Signal denoising involves the manipulation of the signal data to produce a very high quality hearing perception. In this paper we reviews the existing adaptive filter algorithms LMS, NLMS and RLS for de-noising the musical instrument sound signal and performs their comparative study. Here we introduce the variable percentage of additive white Gaussian noise (AWGN) to the sound signal and different adaptive filters are compared. It is observed that RLS algorithm performs better than the other two LMS and NLMS algorithms in terms of peak signal to noise ratio (PSNR), rate of convergence and least time for de-noising the sound signal. Hence this algorithm can also be used for real time applications. It is also observed that the sound signal de-noised with RLS algorithm is very close to the original signal.

Keywords: standard deviation of noise; PSNR; adaptive filtering; RLS; denoising;

Full Text:




  • There are currently no refbacks.

Copyright (c) 2016 International Journal of Advanced Research in Computer Science