phonesensors is a Python package to aid with receiving various sensor data from phones in a quick and easy way. It is
meant to be used in conjunction with the
SensorStreamer
app for Android devices, but support for any other app streaming sensor data over a TCP socket can be implemented
with ease by subclassing the BaseParser in parsers.py.
pip install phonesensors
Open the SensorStreamer app and make it a TCP server emitting JSON packets on a port of your choice. Port 5000 is used for this example. Find out the IP address of your device (e.g. 192.168.1.2). The following code snippet will print the sensor data being streamed from your device:
from phonesensors import SensorStreamerClient
with SensorStreamerClient("192.168.1.2", 5000) as client:
for data in client:
print(data)
For a more detailed guide for how to use PhoneSensors together with the SensorStreamer app, read this article: https://medium.com/p/6d724e3bc427
data in the above example code snippet is a SensorDataCollection instance. You can obtain the specific sensor data
by accessing its attributes. The sensor data and timestamps are returned as numpy arrays:
acceleration_values = packet.acc.values
acceleration_timestamps = packet.acc.timestamps
proximity_values = packet.prox.values
proximity_timestamps = packet.prox.timestamps
Data from different sources may have different number of elements due to differences in sampling frequency. For example,
acceleration_values, a 3D vector, may have 4 samples and thus be a (4,3) array. While proximity_values, a 1D scalar,
may only have one sample and thus be a (1,) array.