iLight: Device-Free Passive Tracking Using Wireless Sensor Networks

In this paper, we study indoor passive tracking problem using wireless sensor networks, in which we assume that a target being tracked is clean, i.e., there is no any equipment carried by the target, hence, the tracking procedure is considered to be passive. We first show that received signal strength indicator and link quality indicator are not as effective as expected for passive detection (tracking) through our extensive testbed studies, and further propose to utilize light to track targets using WSNs. To the best of our knowledge, this is the first work, which studies the passive tracking problem in WSNs using light sensors and general light sources. We present a number of probability-based algorithms to study the moving patterns (properties) of targets being tracked. We design and implement our tracking system named as iLight consisting of 40 wireless sensor nodes and one base station. Through extensive experimental results, we show that iLight can track both single target and multiple targets efficiently.