LoRa ToA-Based Localization Using Fingerprint Method

Localization is one of the essential elements in Internet of Things (IoT) applications for location-based service. In particular, LoRa has suitable features for localization such as low power consumption, long transmission range, and low deployment cost. However, the existing LoRa localization methods have limitations in terms of feasible localization area and accuracy of the localization in noisy environments. In the area where there are less than three gateways, it is difficult to estimate a specific location using the existing algorithms. Also, the existing LoRa localization algorithms result in large estimation error in the noisy environment such as urban. In this paper, we propose a LoRa ToA-based localization using fingerprint map to handle the issues. Through the proposed algorithm, we can estimate location even in the area where there are less than three gateways. In addition, to efficiently calculate distances using ToA, we suggest a LoRa time synchronization protocol (LTSP). By utilizing fingerprint map, the localization error caused by noise and multipath can be significantly reduced. The simulation results show that, compared with other schemes, the proposed algorithm significantly expands feasible localization area while improving the localization accuracy.