High precision indoor positioning algorithm of single LED lamp based on A-Bayes

Abstract

With the rapid development of information technology, the application field of city positioning technology is more and more extensive. Traditional indoor positioning technology has shortcomings in positioning accuracy, equipment cost, security and so on. Visible light indoor positioning technology can effectively solve the above problems. This paper presents an Adaptive boosting- Bayes (A-Bayes) localization algorithm. In this algorithm, three tilted photodetectors and one horizontal PD are used as receivers to measure the received optical power of each PD and set up a fingerprint database. The Adaboost strong classifier is used to divide the test space into two regions: the center and the edge. In the online positioning stage, for the measured received optical power, the region to which it belongs is first determined, and then the Bayesian matching algorithm is used to match the regions to complete the positioning. The experimental results show that in the test space of, and the overall average positioning error is 9.27 cm. Compared with the traditional indoor positioning algorithm, the positioning error of the proposed algorithm is smaller; Compared with the traditional multi lamp location algorithm, the algorithm proposed in this paper avoids the interference between multi light source symbols, and the equipment is simpler; Compared with the traditional fingerprint matching algorithm, the proposed algorithm has faster positioning speed.