Abstract
A new approach based on a histogram analysis to detect QRS-T complexes of the ECG curve rapidly and reliably is described. The aim consists of a rapid and reliable detection of the P-wave (when atrial fibrillation, AF, is absent) or the f-wave (when AF happens). By means of signal processing techniques such as the fractal dimension method and the Shannon entropy method, two types of ECG signals (with or without AF) were analysed and the significant differences between them documented. Accordingly, the methods allow to determine easily whether AF occurs or not.