Cost-Effective Localization Using RSS From Single Wireless Access Point

Abstract

This paper presents a cost-effective localization method by using received signal strength (RSS) from single wireless access point (AP) that has only three directional antennas. To reduce the requirement for the costly database training process, a three-step parametric model-based approach is proposed. This approach estimates the AP-device angle, distance, and the device location analytically by making use of wireless path-loss models, antenna radiation patterns, and differential RSS. Meanwhile, the differential antenna gain is introduced to eliminate the need for known AP-device distances when estimating the AP-device angle. Furthermore, to circumvent the difficulty in detecting outliers in a single-AP system, this paper further integrates single-AP solutions with dead-reckoning (DR) to obtain more robust solutions. Three types of measurement models (i.e., the device location updates, AP-device angle updates, and AP-device distance updates) are built from the single-AP localization system. Static and dynamic field tests with the new Bluetooth 5 signals indicated the possibility for using RSS from three directional antennas on single AP to obtain a low-cost localization solution and the effectiveness of integrating with DR on mitigating the impact of outliers.