| The history of telecardiology dates back to early 1960 when Electrocardiogram (ECG) was delivered via telecommunication transmission lines. Recent advances in wireless telecardiology have brought ease of access, efficiency, remote monitoring, improvement in patient's quality of life and significant reduction in health care costs. The transmission of ECG over wireless communication channel introduces several challenges compared to standard wireline monitoring of cardiovascular activity. In this work, a pre-processing algorithm is presented for reliable recovery of ECG signal corrupted by additive white Gaussian noise (AWGN) even at very low signal-to-noise ratio (SNR) conditions. At low SNR conditions, interference of noise corrupts the ECG signal to such an extent that the received ECG signals when compared to the original signals show unreliable results. The proposed method involves removal of redundancy in the original signal by a two-step approach: transform based compression using orthogonal wavelet basis function followed by entropy encoding. This is to increase the compression ratio but not at the expense of the quality of the reconstructed signal. This is important to preserve the content of clinical information. The aim of the proposed algorithm is to improve the fidelity of the received ECG signal for accurate clinical interpretation. It is done by using data segmentation, data encoding, reassembly, features detection and signal reconstruction. These steps ensure that the diagnostically critical P-wave, QRS complex and T-wave are kept free of large reconstruction errors. The obtained results are evaluated in terms of compression ratio and PRMSD which are around 2.55:1 and 39.62% respectively at 3dB SNR. The visual perception of the reconstructed ECG signal also shows high quality signal recovery. |
