LAUSR

The noise is the most common issue present in all signals. Depending on the process or application, the reduction of the noise is mandatory mainly in areas such as signal classification, pattern recognition, and training process, among others. In recent years, diverse methodologies for noise reduction have been proposed, but they have left some issues with no solution. Such are the noise reduction in signals immersed in white noise without changes in their amplitude or phase or the high performance of recovering components of the desire signals distributed among the frequency spectrum. Also, filtering techniques without significant changes in signal phase and amplitude are necessary. This article proposes a methodology for noise reduction based on the orthogonal matching pursuit algorithm with dictionaries constructed from kernel functions of the discrete cosine transform and discrete sine transform. The methodology is proved in synthetic signals and real electrical signals, reaching noise reduction of white noise (~34 dB in electrical signals) and recovering without distortion.