In this work a non-linear denoising system applied to continuous time signals is presented. The system employs an analog implementation of the semidiscrete wavelet transform (WT). The direct semidiscrete WT is carried out when the input signal is passed through a set of parallel filters, obtaining the wavelet decomposition into various frequency bands. The original signal is recovered by applying the inverse semidiscrete WT. This is carried out by passing the wavelet components through another set of parallel filters and adding the outputs. The system acquires the denoising capacity placing a set of threshold switches, allowing or avoiding the course of the signal components to the sum point. The user adjusts the threshold so that it is greater than the noise component