Analysis of the Position-Dependent Error in FTM RTT Indoor Navigation

Abstract

Fine time measurement (FTM) of the round-trip time (RTT) of a signal between an initiator (smartphone) and a responder (Wi-Fi access point) provides a promising method for indoor positioning. Accurate indoor positioning is a requirement for a wide range of applications, such as asset tracking, indoor navigation, and contact tracing. Unfortunately, the error of reported FTM RTT distance measurements has been shown to have a standard deviation that ranges from 1-2 meters in ideal setups. A major FTM RTT error source was discovered and coined as the “position-dependent error”. This error is heavily depend on the position of an initiator relative to a responder, with the reported measurement fluctuating by meters from an initiator position change of millimeters. Using an Android app and a CNC machine for 2D and 3D positioning, these unusual error properties are explored in depth through experimentation. This experimentation includes evaluating the position-dependent error in both the spatial and frequency domains when varying the test setup, using different smartphones and Wi-Fi access points, and changing the bandwidth and central frequency of the Wi-Fi access points. Possible causes of the position-dependent error are analyzed, such as inaccurate time of arrival or super-resolution algorithms, a dependence on received signal strength, and clock instability. In the end, recommendations for error amelioration are made, and the future of FTM RTT is discussed.