Intelligent beacon location and fingerprinting

Fingerprint Feature Extraction for Indoor Localization

“…This research thus focuses on applying the BLE technology to develop indoor localization methods. Many BLE indoor localization methods [ 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 ] have been proposed in the literature. Among them, BLE fingerprint-based methods have comparably good localization accuracy.…”

“…Five BLE fingerprint-based indoor localization methods [ 16 , 22 , 23 , 24 , 27 ] that are most related to our research are introduced below. Zuo et al [ 16 ] adopted graph optimization to perform indoor localization and produced an error of 1.27 m in the best case.…”

“…Zuo et al [ 16 ] adopted graph optimization to perform indoor localization and produced an error of 1.27 m in the best case. Martins et al [ 22 ] performed indoor localization using Gaussian kernel-based fingerprinting and achieved errors that are less than 1.5 m for approximately 90% of test cases. Subedi et al [ 23 ] employed an improved two-step fingerprint-based localization method, resulting in a localization error of 1.05 m. Li et al [ 24 ] predicted the TD location with an eight-neighborhood template-matching mechanism and achieved a localization error of 1.0 m. Dinh et al [ 27 ] proposed a lightweight and reliable fingerprint-based method using pedestrian dead reckoning and trilateration, bringing about an average error of 0.81 m.…”

“…Experiments are conducted to evaluate FPFE’s performance. FPFE is also compared with the most related methods [ 16 , 22 , 23 , 24 , 27 ] to show its superiority. Furthermore, a practical application using FPFE for smart living is also introduced to show the applicability of FPFE.…”

“…The rest of this paper is organized as follows. Section 2 introduces five fingerprint-based indoor localization methods [ 16 , 22 , 23 , 24 , 27 ] using the BLE technology. The proposed FPFE method is elaborated in Section 3 .…”