Diagnostic approach of reducing computational costs while performing wavelet transformation is reviewed in this paper. Analog-to-digital conversion inevitably produces errors in the process of data measurement, transformation and data processing. It is shown that appliance of fast wavelet algorithms leads to fundamental improvement of effectiveness and increases calculation speed of both according to hardware requirements and minimizing sampling error. Frequency bands analysis showed that for regular signal identification it is not necessary to use complete wavelet decomposition. Usage of lower sampling frequency is enough for accurate signal detection by means of wavelet processing.