| 基于加权质心法的WLAN室内定位系统 |
| |
| 引用本文： | 张勇,徐小龙,徐科宇.基于加权质心法的WLAN室内定位系统[J].电子测量与仪器学报,2015,29(7):1036-1041. |
| |
| 作者姓名： | 张勇 徐小龙 徐科宇 |
| |
| 作者单位： | 1.合肥工业大学计算机与信息学院合肥230009；2.芜湖创业园留学人员博士后科研工作站芜湖241000,合肥工业大学计算机与信息学院合肥230009,合肥工业大学计算机与信息学院合肥230009 |
| |
| 基金项目： | 国家科技支撑计划（2013BAH52F01）、全国大学生创业实践（201410359025）项目 |
| |
| 摘 要： | 在利用接收信号强度指示(RSSI)进行定位的WLAN室内定位系统中,为获得更高的定位精度,提出一种支持向量机与加权质心法相结合的定位算法。该算法首先以四边形对定位场地进行区域划分,在各四边形区域的顶点位置采样指纹点数据,利用支持向量机(SVM)多分类将定位点位置缩小到某个四边形区域内。最后利用加权质心法,计算出定位点的坐标。仿真实验与实地实验结果表明,该算法比支持向量机回归(SVR)、最小二乘支持向量机(LSSVM)、K最近邻法(KNN),定位精度有明显提高,定位误差在1.4 m,定位精度在90%以上。
|
| 关 键 词： | 加权质心法 支持向量机 室内定位 指纹点 |
Algorithm based on weighted centroid method for WLAN indoor positioning |
| |
| Zhang Yong,Xu Xiaolong and Xu Keyu.Algorithm based on weighted centroid method for WLAN indoor positioning[J].Journal of Electronic Measurement and Instrument,2015,29(7):1036-1041. |
| |
| Authors: | Zhang Yong Xu Xiaolong and Xu Keyu |
| |
| Affiliation: | 1.School of Computer and Information, Hefei University of Technology, Hefei 230009, China;2.Post Doctoral Research Center of Wuhu Overseas Students Pioneer Park, Wuhu 241000, China,School of Computer and Information, Hefei University of Technology, Hefei 230009, China and School of Computer and Information, Hefei University of Technology, Hefei 230009, China |
| |
| Abstract: | In order to obtain a higher positioning precision for WLAN indoor positioning system using Received Signal Strength Indicator (RSSI), this paper presents an algorithm based on Support Vector Machines (SVM) and weighted centroid algorithm. The algorithm diverts the located position into quadrilateral regions and samples the RSSI as the fingerprint data in vertex position of the regions. Using multi classification method of SVM, the location of the positioning point is confirmed to one of the quadrilateral regions. Finally, the coordinate of positioning point is computed by using the weighted centroid method. The results of simulation and real environment experiment show that the proposed method has higher accuracy than Support Vector Regression (SVR), Least Squares Support Vector Machine (LSSVM) and K Nearest Neighbor algorithm (KNN). Positioning accuracy is above 90% than other algorithms while the positioning error is 1.4m. |
| |
| Keywords: | weighted centroid algorithm support vector machine indoor positioning fingerprint |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《电子测量与仪器学报》浏览原始摘要信息 |
|
点击此处可从《电子测量与仪器学报》下载全文 |
|
