This paper deals with the noise detection of discrete sEMG signals using wavelets. More specifically, it compares the usefullness of the Haars and Daubechies wavelets for the denoising and compression of the digitalized biosignal. The work is based upon the Discrete Wavelet Transform (DWT) version of Wavelet Package Transform (WPT). A denoising algorithm is proposed to detect unavoidable measured noise in the acquired data. With the help of a seminorm the noise of a sequence is defined. Using this norm it is possible to rearrange the wavelet basis, which can illuminate the differences between the coherent and incoherent parts of the sequence, where incoherent refers to the part of the signal that has either no information or contradictory information. In effect, the procedure looks for the subspace characterised either by small components or by opposing components in the wavelet domain. Concerning the proposed application, an orthosis embedded with a surface electromyography (sEMG) measurement system was used to monitor the electrical activity of the forearm muscles during movement. This method was developed for the monitoring during rehabilitation. The proposed method is general, can be applied to any low frequency signal processing, and was built with wavelet algorithms from the WaveLab 850 library of the Stanford University (USA).