Article Preview
Top1. Introduction
In clinical environment for better diagnosis high resolution noise free Cardiac signal are required. But in practical situations during acquisition, the ECG signal encounters with various types of physiological and non-physiological artifacts. The predominant artifacts present in the ECG includes power line interference (PLI), baseline wander (BW), muscle artifacts (MA) and electrode motion artifacts (EM). In addition to these the tiny features of the ECG signal are masked due to channel noise during the transmission in biotelemetry applications. These artifacts strongly affect the ST segment; degrade the signal quality, frequency resolution, produce large amplitude signals in ECG that can resemble PQRST waveforms and masks tiny features that may be important for clinical monitoring and diagnosis. Cancelation of these artifacts in ECG signals is an important task for better diagnosis. Hence the extraction of high-resolution ECG signals from recordings contaminated with background noise is an important issue to investigate (Ali, 2009). The goal of ECG signal enhancement at the receiving end is to separate the valid signal components from the undesired artifacts and channel noise so as to present an ECG that facilitates easy, accurate interpretation and offering faster diagnosis.
Many approaches have reported in the literature to address ECG enhancement using both adaptive and non-adaptive techniques (Thakor & Zhu, 1991; Kotas, 2006; Ernst, Schlefer, Dieterich, & Schweikard, 2008; Georgy & Dotsinsky, 2008; Hamilton, 1996; Olmos & Laguna, 2000; Poornachandra & Kumaravel, 2006; Rahman, Ahamed, & Reddy, 2011; Rahman, Ahamed, & Reddy, 2012). Adaptive filters permit to detect time varying potentials and to track the dynamic variations of the signals. Besides, they modify their behavior according to the input signal. Therefore, they can detect shape variations in the ensemble and thus they can obtain a better signal estimation. In these papers, the LMS algorithm operates on an instantaneous basis such that the weight vector is updated for every new sample within the occurrence based on an instantaneous gradient estimate. In practical application of adaptive algorithm, a key parameter is the step size. As is well known, if the step size is large, the convergence rate of the LMS algorithm will be rapid, but the steady-state mean square error (MSE) will increase. On the other hand, if the step size is small, the steady state MSE will be small, but the convergence rate will be slow. Thus, the step size provides a trade-off between the convergence rate and the steady-state MSE of the LMS algorithm. An intuitive way to improve the performance of the LMS algorithm is to make the step size variable rather than fixed, that is, choose large step size values during the initial convergence of the LMS algorithm, and use small step size values when the system is close to its steady state, which results in Variable Step Size LMS (VSSLMS) algorithms. Therefore in order to improve convergence rate and filtering capability the step size should be varied continuously in an optimum manner. In literature there are several types of variable step size algorithms are reported (Tyseer & Mayyas, 1997; Lai, 2002; Sulyman & Zerguine, 2003; Shin, Sayed, & Song, 2004; Li, Zhang, Hao, & Chambers, 2008; Vega, Rey, Benesty, & Tressens, 2008; Paleologu, Ciochina, & Benesty, 2008; Zhang, Li, Chambers, & Sayed, 2008; Ni & Li, 2010; Ang & Farhang-Boroujeny, 2001). In clinical situations the noise levels may vary randomly, variable step size strategy is better choice rather than fixed step size. To the best of authors knowledge variable step size algorithms are not used in the removal of non-stationary artifacts in cardiac signals, such an approach results both a fast convergence rate and a small steady-state MSE. Based on this approach various ANCs based on VSSLMS algorithms for ECG enhancement are developed. These are robust variable step-size LMS (RVSSLMS) algorithm providing fast convergence at early stages of adaptation and modified robust variable step-size LMS (MRVSSLMS) algorithm. The performance of these algorithms is compared with conventional LMS and Kowngs VSSLMS algorithm (Kwong & Johnson, 1992). Finally we applied these algorithms on cardiac signal enhancement application. Simulation results confirms that the implemented RVSSLMS and MRVSSLMS are superior than conventional algorithms in terms of convergence rate and signal to noise ratio improvement (SNRI).