Off-the-shelf Wi-Fi Indoor Smartphone Localization

Recently released Wi-Fi adapters, such as Intel AX200 802.11ax NIC, support both Channel State Information (CSI) measurement and Fine Time Measurement (FTM). Angle of Arrival (AoA) estimation with CSI, using MUltiple SIgnal Classification (MUSIC), and FTM are both promising localization methods. But each suffers from practical constraints pertinent to the specific hardware and firmware used. The result can be rather inaccurate localization if AoA or FTM alone were used. We identify the issues/challenges specific to AX200, and as a remedy, we propose a localization approach that combines both CSI-based AoA and FTM. Our approach does not require any modification of the localization target device. This makes the solution readily available for localizing smartphones or any Wi-Fi devices with FTM functionality. Our experimental evaluation shows that our approach achieves a successful localization ratio of 80%, with localization error less than 1m; and less than 0.5m for 66% of the experiments.