Detecting Similar and Dissimilar Movement Patterns using Indoor Passive Localization

Abstract

<p>This paper covers passive localization of a device-free user such that the user is not involved in the positioning system. We use the so-called device-free passive localization approach to estimate the most likely movement pattern exhibited by a remote-controlled, passive robot while moving at a walking speed. Our estimation is based only on the changes in received signal strength as a result of the concerned movement patterns of the robot. This involves observation of physical movement of the robot within an indoor area covered by Wi-Fi signals and its resulting impact on the radio environment. We show that the movement patterns of a remotely controlled robot can be predicted from the signal features.</p>