摘要
传统基于接收信号强度指示(received signal strength indication,RSSI)的室内定位方法易受环境影响且精确度低、稳定性差。为解决这一问题,提出基于受限玻尔兹曼机(restricted Boltzmann machine,RBM)的CSI (channel state information)指纹室内定位方法。在离线阶段时,设计使用RBM来训练校准相应的CSI数据,建立特征指纹库;在线阶段时,利用朴素贝叶斯分类器对数据进行处理并与指纹库匹配,得到定位结果。实验结果表明,基于RBM的CSI指纹室内定位方法优于传统指纹室内定位方法。
The traditional indoor positioning method based on received signal strength(RSSI)is easy to be influenced by the environment with low accuracy and weak stability,a fingerprint based indoor positioning method of channel state information grounded on restricted Boltzmann machine(RBM)was proposed.In the off-line phase,RBM was designed and used to train and adjust the corresponding CSI data,and a feature fingerprint library was built.In the on-line phase,the Naive Bayes Classifier was utilized to process data and match the fingerprint library,and positioning result was obtained.Experimental results show that the method presented is superior to traditional fingerprint based indoor positioning methods.
引文
[1]DENG Zhongliang,YU Yanpei,YUAN Xie,et al.Situation and development tendency of indoor positioning[J].China Communications,2013(3):42-55(in Chinese).[邓中亮,余彦培,袁协,等.室内定位现状与发展趋势研究(英文)[J].中国通信,2013(3):42-55.]
[2]Yang Z,Zhou Z,Liu Y.From RSSI to CSI:Indoor localization via channel response[J].Acm Computing Surveys,2013,46(2):1-32.
[3]LIU Yibo,XIU Chundi,ZHU Yunlong.Indoor positioning method using channel status information[J].Computer Engineering and Applications,2016,52(22):238-241(in Chinese).[刘逸博,修春娣,朱云龙.利用信道状态信息的室内定位方法[J].计算机工程与应用,2016,52(22):238-241.]
[4]Wang X,Gao L,Mao S.CSI phase fingerprinting for indoor localization with a deep learning approach[J].IEEE Internet of Things Journal,2017,3(6):1113-1123.
[5]Kotaru M,Joshi K,Bharadia D,et al.SpotFi:Decimeter level localization using WiFi[J].Acm Sigcomm Computer Communication Review,2015,45(4):269-282.
[6]Wang X,Gao L,Mao S,et al.DeepFi:Deep learning for indoor fingerprinting using channel state information[C]//Wireless Communications and Networking Conference.IEEE,2015:1666-1671.
[7]ZHU Rong,BAI Guangwei,SHEN Hang,et al.Bayesian filtering based CSI indoor localization method[J].Computer Engineering and Design,2015,36(3):567-571(in Chinese).[朱荣,白光伟,沈航,等.基于贝叶斯过滤法的CSI室内定位方法[J].计算机工程与设计,2015,36(3):567-571.]
[8]ZHANG Yanxia.Deep learning models and applications based on the restricted Boltzmann machine[D].Chengdu:University of Electronic Science and Technology of China,2016(in Chinese).[张艳霞.基于受限玻尔兹曼机的深度学习模型及其应用[D].成都:电子科技大学,2016.]
[9]LI Fei,GAO Xiaoguang,WAN Kaifang.Reaesrch on RBMtraining algorithm with dynamic Gibbs sampling[J].ACTAAutomatica Sinica,2016,42(6):931-942(in Chinese).[李飞,高晓光,万开方.基于动态Gibbs采样的RBM训练算法研究[J].自动化学报,2016,42(6):931-942.]
[10]WANG Guocai.Research and application of Naive Bayesian classifier[D].Chongqing:Chongqing Jiaotong University,2010(in Chinese).[王国才.朴素贝叶斯分类器的研究与应用[D].重庆:重庆交通大学,2010.]
[11]WANG Shuangcheng,GAO Rui,DU Ruijie.Dependency extension of Naive Bayesian classifiers based on Gaussian kernel function[J].Control and Decision,2015,30(12):2280-2284(in Chinese).[王双成,高瑞,杜瑞杰.基于高斯核函数的朴素贝叶斯分类器依赖扩展[J].控制与决策,2015,30(12):2280-2284.]
[12]Wang X,Gao L,Mao S.BiLoc:Bi-modal deep learning for indoor localization with commodity 5GHz WiFi[J].IEEEAccess,2017,5(99):4209-4220.
[13]Wang F,Ruan L,Win MZ.Location-aware network operation for cloud radio access network[C]//IEEE International Conference on Acoustics,Speech and Signal Processing.IEEE,2017:3714-3718.
[14]Wang Y,Wu K,Ni LM.WiFall:Device-free fall detection by wireless networks[J].IEEE Transactions on Mobile Computing,2014,16(2):581-594.
