Efficient FIR Filters for Biomedical Signals
Niyan Joseph Savio Marchon · 2019
The most common objectives of the filter is to improve the signal to noise ratio (SNR), disassociate frequency components, obtain the required information from signals for research areas like image, acoustics or bio-medical signal processing. This paper reviews the various types of linear phase FIR filters and proposes a novel linear phase FIR band pass filter design. It was observed that our proposed filter had a flat passband and stopband as compared to the large ripples seen using window based or using other optimal FIR filters. Our simulation synthesis shows that the proposed filter design methodology has reasonably fair transition cut off of 2π for low filter orders. The measure of accuracy for detecting maternal heart rate using the 75 records of the Physionet Challenge dataset closely matched the Physionet Lightwave annotations. Average accuracy for 9 of the 75 records gave less than 100% using N = 101 while, increasing the filter order to N=1001 seven of the nine records increased their accuracy to 100% except for two of the corrupted signals. By re-configuring the filter band edges an efficient band pass filter can be obtained for any biomedical electrical or audio signal.