The fingerprint-based localization technique is one of the most popular indoor localization technologies. There are quite a few localization algorithms that use the RSS distance of position pairs to characterize their physical distance. In this paper, we introduce two coefficients to measure the relationship between RSS distance andphysical distance. Based on the definition of tree-ring distance, we found that the characterization capability of RSS distance to physical distance is closely related to APs’ tree-ring distance. To exploit this, through an in-depth analysis of the relationship between tree-ring distance and physical distance, we pointed out that the APs setscomposed of APs at the edge positions of the positioning area makes the RSS distance better to characterize the physical distance. Further, we proposed a novel RSS distance calculation algorithm based on the comparison of tree-ring distances. In the algorithm, for each pairwiseposition, the abnormal APs are eliminated by the Mean+3S method, and the APs with larger tree-ring distance are selected to participate in the calculation of RSS distance, namely, for different pairwise positions, different APs subsets of all APs are selected to participate in RSS distance calculation. We evaluate the algorithm in a simulation study and initial results show that an APs set with 3 APs is sufficient to guarantee very strong correlation (the correlation coefficient>0.8) and very high consistency (the consistency coefficient>0.8) betweenRSS distance and physical distance, which demonstrates the effectiveness and the practicability of the algorithm.
Bibliographical noteFunding Information:
This work was supported by the National Natural Science Foundation of China under Grant 61702071 and 61501076, High-Level Talents Scientific Research Foundation of Zhoukou Normal University under grant ZKNUC2017024, Guangdong Provincial Key Laboratory of Petrochemical Equipment Fault Diagnosis (Open-ended fund under grant GDUPTKLAB201505) and Key Scientific research projects of colleges and universities in Henan Province under grant 19B520031.
© 2020 Taiwan Academic Network Management Committee. All rights reserved.
Copyright 2020 Elsevier B.V., All rights reserved.
- Fingerprinting localization
- Indoor localization