Communicating is crowdsourcing: Wi-Fi indoor localization with CSI-based speed estimation

Numerous indoor localization techniques have been proposed recently to meet the intensive demand for location-based service (LBS). Among them, the most popular solutions are the Wi-Fi fingerprint-based approaches. The core challenge is to lower the cost of fingerprint site-survey. One of the trends is to collect the piecewise data from clients and establish the radio map in crowdsourcing manner. However the low participation rate blocks the practical use. In this work, we propose a passive crowdsourcing channel state information (CSI) based indoor localization scheme, C²IL. Despite a crowdsourcing based approach, our scheme is totally transparent to the client and the only requirement is to connect to our 802.11n access points (APs). C²IL is built upon an innovative method to accurately estimate the moving speed solely based on 802.11n CSI. Knowing the walking speed of a client and its surrounding APs, a graph matching algorithm is employed to extract the received signal strength (RSS) fingerprints and establish the fingerprint map. For localization phase, we design a trajectory clustering based localization algorithm to provide precise real-time indoor localization and tracking. We develop and deploy a practical working system of C²IL in a large office environment. Extensive evaluations indicate that the error of speed estimation is within 3%, and the localization error is within 2 m at 80% time in a very complex indoor environment.