Towards cognitive device-free localization in wireless networks

Device-free localization (DFL) is a promising technology which could realize localization without equipping the target with a wireless device. It realizes location estimation by monitoring the changes of the radio signal attenuation with a wireless network. One vital problem of the DFL is that its performance will drop significantly when some wireless devices nearby work on the same channel with the DFL system. Motivated by the cognitive radio technique which could realize spectrum sensing efficiently, a novel cognitive DFL (CDFL) scheme is proposed. Different from the traditional DFL system which selects one channel offline and works on this channel online, the CDFL approach could evaluate and detect good channels online and work on the best channel all the time. The experimental results demonstrate that the CDFL could achieve a mean tracking error of only 0.44m, which reduces 52% compared the traditional DFL.