A Machine Learning Approach for Walker Identification Using Smartphone Sensors

Nowadays, smartphones are equipped with MEMS sensors like accelerometers, gyroscopes, and magnetometers. In this work we exploited this kind of sensors to provide advanced information about the walker bringing the smartphone. In particular, smartphone sensors outputs are used to recognize the identity of the walker and the pose of the device during the walk. If the aforementioned information was known, it could be used to improve the functionalities of specific smartphones. For instance, the recognition of walker identity can be used for theft protection or the device pose can be used to improve the performance of the pedestrian navigation. In this paper, we adopted a decision tree classifier approach to recognize the previously described contexts using data produced by smartphone sensors, obtaining effective results.