Comparison between Recursive Least Squares and Optimal Design Methods for Audio Enhancement

This study examines and compares the application of the optimal design method and the recursive least squares (RLS) method to improve the quality of the audio signal in noisy environments. Noise can be incorporated into audio signals through many sources, including amplification systems and electronic switches, which cause loss of signal information or affect the quality of the audio signal. RLS is an adaptive filtering procedure used to design a system that recursively minimizes the noise amplitude of a contaminated signal by comparing the filter output with a desired signal using new incoming signal samples. The optimal design is an FIR filter design technique that has been used to cut parts of the corrupted signal to improve the signal-to-noise ratio. In this study, samples of audio signals contaminated by white noise were used. The noise reduction results show that the RLS approach has vastly improved the quality of the signals. FIR filters, by contrast, can partially improve signal quality. The functionality of the RLS method depended highly on the precision of the measured noise signal. The FIR filter has shown much less signal improvement than the RLS method, but FIR filters are very practical when noise cannot be measured.

Comparison between Recursive Least Squares and Optimal Design Methods for Audio Enhancement Keywords:

Adaptive filtering , error , FIR filter , noise , Optimal Design , RLS filter