Respiration rate calculation using low power DSP processor and SpO2 sensor

This paper proposes a practical implementation of a low power DSP algorithm for respiration rate measurement using the photoplethysmography (PPG) sensor (finger pulse oximeter of SpO₂ sensor). The algorithm was first implemented and tested on a control data using matlab, to prove its validity. A number of datasets were tested with respiration rates ranging from 9 to 39 breaths per minute. The algorithm was ported to the MSP430F5437 DSP microcontroller, and the accuracy, timing and power consumption of the algorithm were measured for different clock frequencies. We showed that an algorithm of such complexity could be executed in real time in an embedded application, and the breath rate could be measured re-using the existing sensor for pulse oximetry, but with lower accuracy than using the dedicated sensor.