Radio Frequency Signature based Implicit Object Movement Detection in an Indoor Environment

Abstract

This paper considers implicit detection problem of object movement in an indoor environment. Specifically, based on radio frequency (RF) attributes in multicarrier transmissions, channel state information (CSI) is exploited to recognize the dynamics of monitoring area. Two wireless communication nodes in the room share the CSI as a byproduct, which fluctuates sensitively according to the movements of objects in the channel space. Thus, by tracking the changes of the CSI, movements of objects can be implicitly detected with simple statistical measures. To this end, we conduct experiments with universal software radio peripheral (USRP) settled onto linear stage in conditions where the object has different sizes and speeds. In numerical evaluation, we analyze the detection performance of the RF signature based implicit object movement detection under various experimental conditions. The results show that the proposed scheme effectively detects the movement of objects, even for small ones less than half of the RF wavelength.