Compression Study of Continuous-Time Sampled ECG Data for e-Health Applications

摘要

The study presented in this paper describes the compressibility of continuous-time sampled electrocardiogram (ECG) signals. The compression method is based on the orthogonal matching pursuit (OMP) algorithm applied to discrete Haar, Daubechie and Biorthogonal wavelets. Comparative results show that Biorthogonal and Daubechie wavelet families offer up to 49% and 43% reductions of the overall number of amplitude samples, respectively. However, Haar wavelets, the less complex wavelet family, discard only 12% of the data and lead to higher computation time than Daubechie family.