Monitoring the heart rate using PPG signals that is worn on the wrist has the downside that it is susceptible to motion. The device on the wrist moves along with the motions of the arm, consequently creating artifacts in the measurement. There are many signal processing technique
...
Monitoring the heart rate using PPG signals that is worn on the wrist has the downside that it is susceptible to motion. The device on the wrist moves along with the motions of the arm, consequently creating artifacts in the measurement. There are many signal processing techniques that can remove these artifacts. This thesis discusses one of the possible ways to remove motional artifacts from PPG signals, which is the TROIKA framework. This framework consists of three parts: signal decomposition, sparse signal reconstruction and spectral peak tracking. The main subject in this thesis is removal of motional artifacts by signal decomposition. Two methods have been proposed to identify components that belong to motional artifacts. Both methods worked as intended for removing those artifacts. The results of the TROIKA framework that is put together were within expectations.
