基于DBSCAN-GRNN-LSSVR算法的WLAN异构终端定位方法
|
摘要
针对WLAN室内定位系统中异构终端(指纹库终端和测试终端)引起的定位偏差过大的问题,提出一种基于DBSCAN-GRNN-LSSVR算法的解决方案。使用最小二乘支持向量回归机(LSSVR)构建指纹库终端接收信号强度(RSS)和物理坐标位置的映射关系模型;列出校准点处异构终端采集的RSS值,得到散点图;用基于密度聚类方法剔除边界点和噪声点;用广义回归神经网络构建异构终端RSS的映射函数;通过LSSVR模型定位测试点的位置。实验结果表明,与只用LSSVR算法相比,测试终端定位精度提高18%～40%,有效解决了定位偏差过大的问题。
Aiming at the excessive location error caused by heterogeneous terminal( fingerprint database terminal and test terminal) in WLAN indoor localization system,this paper proposed a solution based on DBSCAN-GRNN-LSSVR algorithm. This paper employed the least square support vector regression( LSSVR) algorithm to build the mapping relationship model between the received signal strength( RSS) of fingerprint database terminal and physical coordinate locations. It obtained the scatter plot through listing the RSS values collected by the fingerprint database terminal and the test terminal at the calibration point.It eliminated the boundary points and noise points by density-based spatial clustering. It used generalized regression neural network( GRNN) to construct the heterogeneous terminal mapping function of the RSS. It used the LSSVR model to determine the location of the test point. The experiment proves,compared with LSSVR algorithm,using the proposed DBSCAN-GRNNLSSVR algorithm to calibrate the heterogeneous terminal,test terminal positioning accuracy increases by 18% ～ 40%,which effectively solves the problem of excessive localization deviation caused by heterogeneous terminals.
Aiming at the excessive location error caused by heterogeneous terminal( fingerprint database terminal and test terminal) in WLAN indoor localization system,this paper proposed a solution based on DBSCAN-GRNN-LSSVR algorithm. This paper employed the least square support vector regression( LSSVR) algorithm to build the mapping relationship model between the received signal strength( RSS) of fingerprint database terminal and physical coordinate locations. It obtained the scatter plot through listing the RSS values collected by the fingerprint database terminal and the test terminal at the calibration point.It eliminated the boundary points and noise points by density-based spatial clustering. It used generalized regression neural network( GRNN) to construct the heterogeneous terminal mapping function of the RSS. It used the LSSVR model to determine the location of the test point. The experiment proves,compared with LSSVR algorithm,using the proposed DBSCAN-GRNNLSSVR algorithm to calibrate the heterogeneous terminal,test terminal positioning accuracy increases by 18% ～ 40%,which effectively solves the problem of excessive localization deviation caused by heterogeneous terminals.
引文
[1]张明华,张申生,曹健.无线局域网中信号强度的室内定位[J].计算机科学,2007,34(6):68-71,75.(Zhang Minghua,Zhang Shensheng,Cao Jian.Received-signal strength-based indoor location in wireless LANS[J].Computer Science,2007,34(6):68-71,75.)
[2]秦泗明.基于位置指纹的Wi-Fi室内定位技术研究[D].成都:电子科技大学,2013.
[3]Dong Fangfang,Chen Yiqiang,Liu Junfa,et al.A calibration-free localization solution for handling signal strength variance[C]//Proc of International Conference on Mobile Entity Localization and Tracking in GPS-Less Environments.Berlin:Springer-Verlag,2009:79-90.
[4]Kim Y,Shin H,Chon Y,et al.Smartphone-based Wi-Fi tracking system exploiting the RSS peak to overcome the RSS variance problem[J].Pervasive&Mobile Computing,2013,9(3):406-420.
[5]Kjaergaard M B,Munk C V.Hyperbolic location fingerprinting:a calibration-free solution for handling differences in signal strength[C]//Proc of IEEE International Conference on Pervasive Computing.Piscataway,NJ:IEEE Press,2008:110-116.
[6]赵聪.基于位置指纹的WLAN室内定位算法研究[D].哈尔滨:哈尔滨工业大学,2014.
[7]Hossain A K M M,Jin Yunye,Soh W S,et al.SSD:a robust RF location fingerprint addressing mobile devices’heterogeneity[J].IEEETrans on Mobile Computing,2013,12(1):65-77.
[8]谷洋,蒋鑫龙,刘军发,等.设备自适应的无线信号特征提取与定位方法[J].软件学报,2014,25(S2):12-20.(Gu Yang,Jiang Xinlong,Liu Junfa,et al.Device adaptive wireless signal feature extraction and localization method[J].Journal of Software,2014,25(S2):12-20.)
[9]罗清旺,师奕兵,王志刚,等.铁磁性管道物理参数反演方法研究[J].仪器仪表学报,2016,37(1):9-14.(Luo Qingwang,Shi Yibing,Wang Zhigang,et al.Research on the inversion of physical parameters of the ferromagnetic pipe[J].Chinese Journal of Scientific Instrument,2016,37(1):9-14.)
[10]代亮,许宏科,陈婷,等.基于MapReduce的最小二乘支持向量机回归模型[J].计算机应用研究,2015,32(4):1060-1064.(Dai Liang,Xu Hongke,Chen Ting,et al.Least squares support vector machine regression model based on MapReduce[J].Application Research of Computers,2015,32(4):1060-1064.)
[11]Suykens J,Van G T,De B P,et al.Least square support sector machines[M].Singapore:World Scientific Publishing,2002:71-111.
[12]田增山,王向勇,周牧,等.基于DBSCAN子空间匹配的蜂窝网室内指纹定位算法[J].电子与信息学报,2017,39(5):1157-1163.(Tian Zengshan,Wang Xiangyong,Zhou Mu,et al.DBSCAN based subspace matching for indoor cellular network fingerprint positioning algorithm[J].Journal of Electronics&Information Technology,2017,39(5):1157-1163.)
[13]冯振华,钱雪忠,赵娜娜.Greedy DBSCAN:一种针对多密度聚类的DBSCAN改进算法[J].计算机应用研究,2016,33(9):2693-2700.(Feng Zhenhua,Qian Xuezhong,Zhao Na’na.Greedy DB-SCAN:an improved DBSCAN algorithm on multi-density clustering[J].Application Research of Computers,2016,33(9):2693-2700.)
[14]郭保青,朱力强,史红梅.基于快速DBSCAN聚类的铁路异物侵限检测算法[J].仪器仪表学报,2012,33(2):241-247.(Guo Baoqing,Zhu Liqiang,Shi Hongmei.Intrusion detection algorithm for railway clearance with rapid DBSCAN clustering[J].Chinese Journal of Scientific Instrument,2012,33(2):241-247.)
[15]吴常铖,宋爱国,曾洪,等.基于s EMG和GRNN的手部输出力估计[J].仪器仪表学报,2017,38(1):97-104.(Wu Changcheng,Song Aiguo,Zeng Hong,et al.Force estimation based on s EMG and GRNN[J].Chinese Journal of Scientific Instrument,2017,38(1):97-104.)
[16]葛柳飞,李克清,戴欢.基于自适应GRNN的无线室内定位算法[J].计算机工程,2016,42(6):81-85,90.(Ge Liufei,Li Keqing,Dai Huan.Wireless indoor location algorithm based on adaptive GRNN[J].Computer Engineering,2016,42(6):81-85,90.)
[2]秦泗明.基于位置指纹的Wi-Fi室内定位技术研究[D].成都:电子科技大学,2013.
[3]Dong Fangfang,Chen Yiqiang,Liu Junfa,et al.A calibration-free localization solution for handling signal strength variance[C]//Proc of International Conference on Mobile Entity Localization and Tracking in GPS-Less Environments.Berlin:Springer-Verlag,2009:79-90.
[4]Kim Y,Shin H,Chon Y,et al.Smartphone-based Wi-Fi tracking system exploiting the RSS peak to overcome the RSS variance problem[J].Pervasive&Mobile Computing,2013,9(3):406-420.
[5]Kjaergaard M B,Munk C V.Hyperbolic location fingerprinting:a calibration-free solution for handling differences in signal strength[C]//Proc of IEEE International Conference on Pervasive Computing.Piscataway,NJ:IEEE Press,2008:110-116.
[6]赵聪.基于位置指纹的WLAN室内定位算法研究[D].哈尔滨:哈尔滨工业大学,2014.
[7]Hossain A K M M,Jin Yunye,Soh W S,et al.SSD:a robust RF location fingerprint addressing mobile devices’heterogeneity[J].IEEETrans on Mobile Computing,2013,12(1):65-77.
[8]谷洋,蒋鑫龙,刘军发,等.设备自适应的无线信号特征提取与定位方法[J].软件学报,2014,25(S2):12-20.(Gu Yang,Jiang Xinlong,Liu Junfa,et al.Device adaptive wireless signal feature extraction and localization method[J].Journal of Software,2014,25(S2):12-20.)
[9]罗清旺,师奕兵,王志刚,等.铁磁性管道物理参数反演方法研究[J].仪器仪表学报,2016,37(1):9-14.(Luo Qingwang,Shi Yibing,Wang Zhigang,et al.Research on the inversion of physical parameters of the ferromagnetic pipe[J].Chinese Journal of Scientific Instrument,2016,37(1):9-14.)
[10]代亮,许宏科,陈婷,等.基于MapReduce的最小二乘支持向量机回归模型[J].计算机应用研究,2015,32(4):1060-1064.(Dai Liang,Xu Hongke,Chen Ting,et al.Least squares support vector machine regression model based on MapReduce[J].Application Research of Computers,2015,32(4):1060-1064.)
[11]Suykens J,Van G T,De B P,et al.Least square support sector machines[M].Singapore:World Scientific Publishing,2002:71-111.
[12]田增山,王向勇,周牧,等.基于DBSCAN子空间匹配的蜂窝网室内指纹定位算法[J].电子与信息学报,2017,39(5):1157-1163.(Tian Zengshan,Wang Xiangyong,Zhou Mu,et al.DBSCAN based subspace matching for indoor cellular network fingerprint positioning algorithm[J].Journal of Electronics&Information Technology,2017,39(5):1157-1163.)
[13]冯振华,钱雪忠,赵娜娜.Greedy DBSCAN:一种针对多密度聚类的DBSCAN改进算法[J].计算机应用研究,2016,33(9):2693-2700.(Feng Zhenhua,Qian Xuezhong,Zhao Na’na.Greedy DB-SCAN:an improved DBSCAN algorithm on multi-density clustering[J].Application Research of Computers,2016,33(9):2693-2700.)
[14]郭保青,朱力强,史红梅.基于快速DBSCAN聚类的铁路异物侵限检测算法[J].仪器仪表学报,2012,33(2):241-247.(Guo Baoqing,Zhu Liqiang,Shi Hongmei.Intrusion detection algorithm for railway clearance with rapid DBSCAN clustering[J].Chinese Journal of Scientific Instrument,2012,33(2):241-247.)
[15]吴常铖,宋爱国,曾洪,等.基于s EMG和GRNN的手部输出力估计[J].仪器仪表学报,2017,38(1):97-104.(Wu Changcheng,Song Aiguo,Zeng Hong,et al.Force estimation based on s EMG and GRNN[J].Chinese Journal of Scientific Instrument,2017,38(1):97-104.)
[16]葛柳飞,李克清,戴欢.基于自适应GRNN的无线室内定位算法[J].计算机工程,2016,42(6):81-85,90.(Ge Liufei,Li Keqing,Dai Huan.Wireless indoor location algorithm based on adaptive GRNN[J].Computer Engineering,2016,42(6):81-85,90.)