Joint reduction of baseline wander, PLI and its harmonics in ECG signal using Ramanujan Periodic Transform

Ramanujan Periodic Transform (RPT) is a newly emerging transformation technique in the field of signal processing. It uses an integer bases (obtained from Ramanujan sum) for transformation. A recorded ECG signal often contains artifacts (bioelectric signals) namely, baseline wander, muscle artifacts (EMG-Electromyogram), motion artifacts, powerline interference (PLI) and its harmonics. With certain precautions during signal recording we can avoid both muscle and motion artifacts. The other noises can be reduced by preprocessing of the recorded ECG signal. In this paper, RPT is used for preprocessing, to reduce baseline wander noise, PLI and its harmonics. The proposed methodology is tested on a record from MIT-BIH Arrhythmia database for different block sizes. A sum (E) of Euclidean errors per block (e
_i
-i
^th
block), is used as a measure to compare the results of RPT with notch filter technique. From the results, the RPT is reducing the noise with minimum error (E), when compared with notch filter technique.