Aiming at the complex indoor environment, a new indoor localization algorithm was proposed based on the K-nearest neighbour algorithm (KNN) and extreme learning machine (ELM) by using virtual reference tags. In this paper, LANDMARC location algorithm is used in regional location during the online phase, and the ELM with virtual reference tags was introduced in the locked area. The design scheme of indoor positioning system based on Intel R1000 platform is proposed, and the system was realized by C++ language. The experimental data show that average error of the indoor positioning system is 0.3m and the reduction in estimation error is 38% over LANDMARC and 19% over ELM. The system can effectively improve the indoor localization accuracy in low tag density environments.
RFID Localization System Based on K-Nearest Neighbor Algorithm and Extreme Learning Machine Algorithm with Virtual Reference Tags
2019-04-01
American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS); Vol 54 No 1 (2019); 82-91 ; 2313-4402 ; 2313-4410
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
DDC: | 629 |
Fast nearest-neighbor search algorithm [2669-11]
British Library Conference Proceedings | 1996
|Fast multidimensional nearest neighbor search algorithm using priority queue
Online Contents | 2008
|Attack identification method based on K nearest neighbor algorithm and charging device
Europäisches Patentamt | 2020
|Fast Algorithm for Nearest Neighbor Search Based on a Lower Bound Tree
British Library Conference Proceedings | 2001
|