Lighthouse: Precise 802.11-based Localization

This paper proposes a novel approach to the problem of indoor location estimation based on IEEE 802.11 (WiFi) signals. The approach uses a rotating directional antenna at the Access Point (AP). The Mobile Device (MD) measures the variation of Received Signal Strength Indication (RSSI) with antenna Direction Of Emission (DOE). The observed DOE-RSSI signature is used as a fingerprint to search a map of previously measured DOE-RSSI signatures. The method has the advantage that it works with existing 802.11 cards and that it provides higher accuracy, for the same number of APs, than the conventional RSSI fingerprinting technique based on maps obtained using multiple omnidirectional APs. We report results obtained from an experimental testbed deployed in a typical office space. The Lighthouse algorithm achieved 85% and 95% success ratio in MD localisation over 20 possible positions using one AP and two APs, respectively. The success ratio achieved is, on average, +24% and +53% better than the conventional RSSI fingerprinting and directional beaconing methods, respectively.