期刊名称：International Journal of Future Generation Communication and Networking
印刷版ISSN：2233-7857
出版年度：2016
卷号：9
期号：3
页码：201-214
DOI：10.14257/ijfgcn.2016.9.3.19
出版社：SERSC
摘要：Against the problem that indoor positioning suffers quite large errors and irregular user location movement, this paper adopts Support Vector Regression (SVR) for initial positioning and Kalman filtering for filtering of the positioning results so as to improve the accuracy of the positioning system. The experimental results show that against the real WLAN environment, SVR positioning results processed by Kalman filtering indicates the root mean square error is decreased by 16%, and 73% of positioning accuracy within 2 meters is increased to 83%.
