Oversampled Graph Filter Banks For Acoustic Signal Processing

 Graph signal processing must explicitly consider the structure of the signal. To propose M channel filter bank for graph signal. Oversampled graph filter bank can be used to de-noise graph signals. Over sampling improves resolution, reduces noise and helps avoiding aliasing and phase distortion by relaxing anti-aliasing filter performance. Oversampled scheme that uses an oversampled wavelet transform in oversampled scheme. It provides better noiseless graph signal. Here noise reduction can be done by using Chebychev filter. Chevebychev filter minimize the error between the idealized and the actual filter characteristic over the range of the filter. Wavelet transform captures both frequency and time.