Open AccessA Modified KNN Indoor WiFi Localization Method With K-median Cluster
Author(s) -
Lan Wei,
Hongxin Li
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/608/1/012008
Subject(s) - robustness (evolution) , computer science , cluster analysis , k nearest neighbors algorithm , global positioning system , signal strength , fingerprint (computing) , artificial intelligence , similarity (geometry) , pattern recognition (psychology) , algorithm , data mining , wireless , telecommunications , biochemistry , chemistry , image (mathematics) , gene
Because of the serious attenuation and multi-path effect of GPS signal, outdoor-positioning technology can not be applied in complex indoor environment. Through the study of K-Nearest Neighbor applied in WiFi positioning, according to the problem that the time complexity of KNN algorithm increases linearly with the quantity of samples, this paper combined clustering algorithm with KNN optimized the similarity measure in fingerprint feature space and proposed a efficient indoor target location algorithm . Experimental results showed that the algorithm improved the positioning accuracy, had strong robustness to noise and more importantly, the positioning time was effectively shortened and it can meet the requirements of practical applications.