Research Article
MTAPS: Indoor Localization Algorithm Based on Multiple Times AP
@INPROCEEDINGS{10.1007/978-3-030-32216-8_2, author={Pengyu Huang and Haojie Zhao and Wei Liu and Dingde Jiang}, title={MTAPS: Indoor Localization Algorithm Based on Multiple Times AP}, proceedings={Simulation Tools and Techniques. 11th International Conference, SIMUtools 2019, Chengdu, China, July 8--10, 2019, Proceedings}, proceedings_a={SIMUTOOLS}, year={2019}, month={10}, keywords={Location fingerprint Multiple access point selections K-means Location clusters Decision tree}, doi={10.1007/978-3-030-32216-8_2} }- Pengyu Huang
Haojie Zhao
Wei Liu
Dingde Jiang
Year: 2019
MTAPS: Indoor Localization Algorithm Based on Multiple Times AP
SIMUTOOLS
Springer
DOI: 10.1007/978-3-030-32216-8_2
Abstract
In recent years, indoor localization base on fingerprint has become more and more common. In many fingerprint-based indoor positioning algorithms, it’s very popular to use WiFi signal characteristics to represent the location fingerprint. However, with the great improvement of IEEE 802.11 protocols, WiFi has been broadly used. So there are numbers of WiFi access points (APs) have been deployed everywhere which can be used for localization purpose. The large amount of AP can greatly increase the dimension of the fingerprint and localization complexity. In this paper, we propose a novel indoor positioning algorithm MTAPS (indoor localization algorithm based on multiple times access point selection). MTAPS can effectively reduce the complexity of localization computation, and improve the performance of localization with an efficient access point selection algorithm. This indoor localization algorithm can get a better subset of APs through multiple times AP selection method. These selected APs will be more stable and can provide a better discriminative capability to reference locations. In addition, MTAPS uses k-means algorithm to cluster reference locations, and makes up a decision tree for every location cluster. After location clustering, MTAPS re-selects a suitable AP subset for every cluster. This method can further improve localization performance. Experimental results show that MTAPS has better localization performance than the indoor localization algorithm which is based on classical AP selection algorithm. And MTAPS can achieve the accuracy of over 90% within 2 m localization error.
