Software Campus: AKTIFUNK

July 24th, 2014  |  Published in Research

Human activity recognition, i.e. the recognition of a users’ current physical activity (e.g. walking, eating, sleeping) by monitoring and analyzing her/his behavior and the environment, can improve quality
of living. However, current sensors for activity recognition have a number of drawbacks such as installation/attachment effort, battery runtime or privacy concerns.

In AKTIFUNK, TECO is researching the use of radio signal changes to derive user activities. As this approach may offer an alternative recognition possibility without the mentioned drawbacks. Here the user is not outfitted with a device. Instead the existing wireless infrastructure (e.g. WiFi, GSM, AM/FM Radio, Bluetooth, ZigBee) is concurrently used to determine the human physical activity of the user.

More specifically, AKTIFUNK aims to investigate propagation impact of users on the radio signals, develop novel recognition algorithms and evaluate these algorithms regarding practical online recogniton and long-term robustness.

AKTIFUNK

START/END

  • 01/2014 – 12/2014

PARTNERS

  • Robert Bosch GmbH

RESEARCH TOPICS

  • Human Activity Recognition
  • Wave Propagation
  • Device-Free, Radio-Based Context Recognition

CONTACT

Selected Publications

Markus Scholz, Lukas Kohout, Matthias Horne, Matthias Budde, Michael Beigl, Moustafa Youssef (2015) Device-Free Radio-based Low Overhead Identification of Subject Classes, 2nd Workshop on Physical Analytics (WPA-15), co-located with ACM Mobisys

Tags: ,

Comments are closed.