Key result
RLS adaptive filtering improves ECG denoising with faster convergence and higher SNR versus LMS.
Why the study?
ECG signals are commonly corrupted by physiological and non-physiological noises, and adaptive filters are important for noise cancellation but LMS algorithm has a low convergence rate.
The RLS adaptive algorithm provides superior noise cancellation and faster convergence for ECG signal processing in telecardiology compared to traditional LMS algorithms.
When acquiring the Electrocardiogram (ECG) signal from the person, it should be preprocess before sending to the analyst for taking decision of the signal, because signal should be affected with various artifacts.For numerous applications of noise cancellation in the corrupted signals, adaptive filters play important role.The various artifacts which commonly occur in the acquisition of ECG signals are physiological and non-physiological noises, those are main supply power line interference, muscle artifact, electrode motion artifact and base line wander noises.The adaptive Least Mean Square (LMS) algorithm provides a low convergence rate, so that for fast convergence rate and reduced noise, in this paper an efficient Recursive Least Square algorithm is considered, for removing of power line noise and muscle noise.For double validation of the signal, and for high Signal to Noise Ratio (SNR), fast convergence rate, is achieved by using LMS to RLS adaptive algorithm at the cost of additional computations.
No takes yet. Share an insight, caveat, or question.
Gowri et al. (2015) studied ECG signal noise. Recursive Least Square (RLS) adaptive algorithm vs. Least Mean Square (LMS) algorithm was evaluated on Signal to Noise Ratio (SNR) and convergence rate. An efficient Recursive Least Square (RLS) adaptive algorithm achieved a high Signal to Noise Ratio and fast convergence rate for removing power line and muscle noise from ECG signals.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: