Dataset from PPG wireless sensor for activity monitoring

Area of Science:

• Biomedical Engineering
• Signal Processing
• Machine Learning

Background:

• Photoplethysmography (PPG) signals are widely used for physiological monitoring.
• Motion artifacts significantly degrade PPG signal quality, especially from wrist-worn devices.
• Simultaneous accelerometer data is essential for motion artifact removal and activity recognition.

Purpose of the Study:

• To introduce a novel dataset of simultaneously acquired wrist-based PPG and tri-axial accelerometer signals.
• To provide data for developing and validating algorithms for human activity recognition.
• To facilitate research on PPG signal quality under motion conditions.

Main Methods:

• Data collected from 7 subjects.
• Acquired 105 PPG signals and 105 corresponding tri-axial accelerometer signals.
• Sampling frequency of 400 Hz for all signals.

Main Results:

• A comprehensive dataset containing synchronized PPG and accelerometer signals is presented.
• The dataset captures PPG signals under realistic motion artifact conditions.
• The data is suitable for training and testing machine learning models.

Conclusions:

• The introduced dataset offers valuable resources for advancing human activity recognition.
• This data enables the development of robust algorithms capable of handling motion artifacts in PPG signals.
• The dataset supports research in wearable sensor technology and physiological monitoring.