Respiratory Rate Measuring System Using Accelerometer and Gyroscope Sensor in Android Smartphone With Variational Mode Decomposition (VMD)

Measurement of respiration rate is a critical parameter in understanding human breathing patterns. Conventional medical devices for respiration measurement are often expensive, require special setups, and lack portability. This study aims to develop a respiration rate measurement system based on Android smartphones, utilizing built-in accelerometer and gyroscope sensors as an alternative solution. The system is designed for use in a supine position, with the device placed on the diaphragm area to detect breathing patterns. The research dataset consists of 10 subjects, each undergoing 30 tests, yielding a total of 300 data samples. Data were recorded from various combinations of device positions (Landscape and Portrait) and diaphragm locations (Right, Center, and Left). The data obtained were processed through several stages: normalization, filtering using a median filter, signal decomposition using Variational Mode Decomposition (VMD), and peak detection to calculate the respiration rate per minute (RPM). Analysis results revealed that the primary respiration signal was identified in IMF 2, with dominant frequencies in the range of 0.2–0.3 Hz. Testing demonstrated an average accuracy of 95.19% and precision of 98.45%, with deviations of less than 2% compared to manual measurements. The combination of Portrait position and Center diaphragm location produced the most stable and accurate measurements. This study offers a simple solution for respiration measurement using smartphone-based devices that can be utilized for further research or to support the development of portable health monitoring tools.