Modified Model of RLS Adaptive Filter for Noise Cancellation
Nilesh Kumar Yadav,Amit Dhawan,Manish Tiwari,Sumit Kumar Jha
DOI: https://doi.org/10.1007/s00034-024-02605-5
IF: 2.311
2024-02-13
Circuits Systems and Signal Processing
Abstract:Recursive Least Square (RLS) is a popular algorithm for noise cancellation in non-stationary signals; however, it demands more computational resources and more difficult mathematical operations. Also, RLS has less performance stability. The present research explores a novel idea of an RLS adaptive noise cancellation through a modified method that uses an RLS adaptive filter by introducing an additional constant multiplier, along with their in-depth analysis. The proposed algorithm is analyzed using three primary performance metrics: mean square error (MSE), signal-to-noise ratio (SNR), and convergence rate. The obtained results demonstrate that the proposed algorithm has reduced the MSE by 79.65%, which leads to an improvement in SNR by 86.16% compared to the traditional RLS algorithm. The additional constant multiplier is optimized for SNR, and the optimized value is found to be equal to 0.65, which gives the best possible SNR value of 19.38 dB. Also, the proposed algorithm has been successfully applied to a real-world scenario in acoustic echo cancellation (AEC). The experimental setup for the echo canceller is simulated on MATLAB to measure the echo canceller efficiency in terms of MSE and echo return loss enhancement. Based on performance evaluation, the proposed algorithm has been found to better echo cancellation in AEC as compared to traditional RLS.
engineering, electrical & electronic