For a full description with methodology, results, and discussion; please see PDF report in the repository or, alternatively, in https://tinyurl.com/y5pt3u7w.
The dataset contains data from accelerometers, EMG sensors, and ECG sensors.
Accelerometer data collected using the Phyphox application on an iOS iPhone. The data was captured without gravity (g) meaning that only movements were recorded. The timestamps in the dataset are not constant, as data was written only when participants made movements on their hips or legs. This resulted in timestamps being irregular, occurring only when significant movements were detected.
EMG data was collected from the participants' dominant arm using Trigno Quattro Sensors and the Trigno EMG Reports Android app. It is available in both formats (.xlsx and .shpf), while curated data is available in .csv format. Sensor postions on the arms can be consulted in the report. Note that for some participants C/D sensors were switched. The affected data however has been reported in participants_data.csv and has no integrity or reliability problems other than labelling.
ECG data was collected with Shimmer sensors and the acompaning Windows software. Sensor postions on the arms can be consulted in the report.
The dataset includes data from 20 participants, each identified as P1 to P20. The duration of data collected from each participant varies based on their movements. The data was recorded from two pockets, left and right, on the participants' clothing. It is important to note that the durations of data collected from the accelerometers left and right pockets may not always be equal. This discrepancy in duration occurs because the data recording was manually stopped by the tester after the actions or movements being recorded ceased. Therefore, there may be differences in the durations of data collected from the left and right pockets, even for the same participant (P).