高技术通讯（英文版）2010，Vol.16Issue(3)：227-234,8.DOI:10.3772/j.issn.1006-6748.2010.03.002
WLAN indoor location method based on artificial neural network
WLAN indoor location method based on artificial neural network
摘要
Abstract
WLAN indoor location method based on artificial neural network (ANN) is analyzed.A three layer feed-forward ANN model offers the benefits of reducing time cost of the layout of an indoor location system, saving storage cost of the radio map establishment and enhancing real-time capacity in the on-line phase.According to the analysis of SNR distributions of recorded beacon signal samples and discussion about the multi-mode phenomenon, the one map method is proposed for the purpose of simplifying ANN input values and increasing location performances.Based on the simulations and comparison analysis with other two typical indoor location methods, K-nearest neighbor (KNN) and probability, the feasibility and effectiveness of ANN-based indoor location method are verified with average location error of 2.37m and location accuracy of 78.6% in 3m.关键词
indoor location/WLAN/artificial neural network (ANN)/multi-mode/fingerprintKey words
indoor location/WLAN/artificial neural network (ANN)/multi-mode/fingerprint分类
计算机与自动化引用本文复制引用
Zhou Mu ,Sun Ying,Xu Yubin,Deng Zhian,Meng Weixiao..WLAN indoor location method based on artificial neural network[J].高技术通讯（英文版）,2010,16(3):227-234,8.基金项目
Supported by the National High Technology Research and Development Programme of China (No.2008AA12Z305). （No.2008AA12Z305）
