Multichannel ECG data compression method based on a new modeling method

This work addresses the problem of multichannel ECG data compression. The proposed method is based on the identification of an FIR system. The idea of this method is to consider different independent leads (I, II, V1 to V6) as the outputs of linear filters in which the excitation signal is one of these leads suitably chosen. For purposes of compression, the parameters of the FIR system are quantized and coded. Compression results are presented in terms of the signal-to-noise ratio (SNR) and compression ratio (CR). The results obtained shown a good reconstruction level with a high CR.