Performance of variable step size LMS adaptive algorithm for the removal of artifacts from electrocardiogram using DSP processor

摘要

Nowadays, telecardiology becomes an important tool in cardiac diagnosis from a remote location. During Electrocardiogram (ECG) acquisition, several artifacts strongly degrade the signal quality and mask the tiny features which are important for Clinical monitoring and diagnosis. In some emergency cases the ECG signal has to be transmitted to the clinic from the ambulance. During the transmission the tiny features of the ECG signal are masked due to channel noise. In practice, most of the signals are non-stationary and fixed coefficients filters will not work well. From the past history it was proven that among several filtering techniques, adaptive filters can efficiently adjust its coefficients in accordance to statistical nature of the input signal. The development of efficient adaptive noise cancellers to remove artifacts from ECG signals would therefore be very beneficial. The reduction of complexity in telecardioalogy system applications is important. In this paper we are going to present various less computational complexity adaptive filtering methods based on variable step-size algorithms. The main aim is to eliminate the affect of PLI artifact using these adaptive algorithms and performance of simulated result is compared with MATLAB and DSP processor kit. Simulation results shows that the proposed realization gives better performance in terms of signal to noise ratio.