Activity recognition from RSSI on android mobile phones

Details

Supervisor: Stephan Sigg
Duration: 3-6 months
Type: Bachelor/Master Thesis or Student Project
Status: open


Project Description

Recognition of activities on mobile phones is typically conditioned on inertial sensors such as accelerometers. However, the phone is, although in the same room as its owner about 90% of the time, within arms reach only about 50% of a day (cf. [1],[2]). Consequently, inertial sensors can not be utilised for activity recognition most of the time since the phone is discarded. However, we have recently demonstrated the recognition of activities and gestures based on RSSI captured on a mobile phone. With RSSI information, the perception of a phone can be extended to its environments.

In this project, the student(s) will optimise and implement activity recognition from RSSI fluctuation on a mobile phone. There are various questions to be solved in this topic. The exact task will therefore be defined in cooperation with the student and depend on the type of thesis/project conducted. We will provide the hardware required and preliminary implementations which can be used as a starting point.

Required Skills

  • Natural curiosity, high motivation and a good sense of creativity :)
  • Programming experience with android is beneficial but can also be acquired in the course of the project