DETECTION OF ECG SIGNALS USING WAVELET BASED ECG DETECTOR

Abstract

Recent researches are focused on monitoring and processing of biomedical signals. Electrocardiogram (ECG) is a non- invasive technique which plays a very important role in detection and diagnosis of such signals. ECG is the recording of a signal which interprets the electrical activity of the heart over a period of time, which is detected by electrodes attached to the surface of the skin. This electrical activity is demonstrated by certain changes of electrical voltage on the body surface. And these changes are recorded by summarizing the electrical activity of all heart cells. The digital signal processing algorithms which analyze ECG signals are mainly divided in three categories: time domain, frequency domain and time- frequency domain. The algorithm using time- frequency domain analysis overcomes the drawbacks suffered by earlier two methods. The time domain algorithms do not provide high quality results whereas, the frequency domain algorithms cannot specify the exact phase of the heart cycle during which changes take place. The ECG signal is basically a non- stationary signal, as it changes its physiological and statistical property with respect to time and so wavelet transforms are a useful tool to study such types of signals. Wavelet Transform (WT) utilizes time- frequency analysis which is used for detection of ECG signals and their implementation. In this work, the conventional wavelet based ECG detector has been studied in detail and simulated using MATLAB (R2012b). The design for detection of ECG signals has been extended to employ multiwavelet approach. This proposed ECG detector architecture has been simulated and verified in terms of its workability.