Analysis of Finite Impulse Response (FIR) Filter to Reduce Motion Artifacts of Heart Rate Signal Based on Photoplethysmography

Currently, heart rate monitoring method using photoplethysmography has attracted a lot of attention because of its advantages. However, the results of photoplethysmography (PPG) recordings are susceptible to interference with motion artifacts (MA) that can reduce the accuracy of their readings. This study aimed to determine the performance of the Finite Impulse Response (FIR) filter by analyzing the filter type and order in reducing the motion artifacts of the photoplethysmography heart rate signal. This research was conducted using SKU sensors: SEN0203 to get PPG signal because it has specifications to be used for reading heart rate signal, Arduino Nano, and MATLAB to design FIR filter. Furthermore, FFT and PSD which consist of frequency-based features and use skewness value were also employed in order to know the performance of the filter. The study was conducted on 10 respondents in their relax condition and moving condition in the forms of bending finger. The results showed that mean frequency and skewness value had a strong correlation to the PPG signal. By using a filter, the mean frequency error value could be reduced from 10 to 0.23% and the error skewness value could be reduced from 74.9 to 21.69%. Based on several performances, we concluded that FIR filter least square on order 50 could reduce MA with good performance among the other filter that we have been tested. This study is useful as a framework for finding out the right features in the PPG signal classification method and selecting the appropriate type of filter for future application.