An Implementation in dsPIC of a Denoising Algorithm Based on the Discrete Wavelet Transform

Abstract

In speech processing, one of the first steps is to reduce the noise that the signal contains, for this purpose there are several algorithms. Actually, the most widely used noise reduction methods are based on the discrete wavelet transform (DWT). The main advantages of these methods are that they work with non-stationary signals and have the ability to perform a multi-resolution analysis. In counterpart, its main disadvantage is that the amount of required data memory difficult their implementation in a low cost digital signal processor (DSP). The main goal of this work is to implement a noise reduction algorithm based on the DWT in a dsPIC of Microchip. Experimental measurements are carried out to analyze the performance of the system.