《计算机应用研究》|Application Research of Computers

基于可靠AP选择和深度置信网络的室内定位算法

Indoor positioning algorithm based on reliable AP selection and deep belief network

免费全文下载 (已被下载 次)  
获取PDF全文
作者 李新春,郭欣欣
机构 辽宁工程技术大学 a.电子与信息工程学院;b.研究生院,辽宁 葫芦岛 125105
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2018)08-2469-05
DOI 10.3969/j.issn.1001-3695.2018.08.058
摘要 受非视距传播等影响,基于位置指纹的室内定位精度不高。针对此问题,提出一种基于可靠AP选择和深度置信网络(DBN)的室内定位算法。离线阶段利用改进K-means算法将定位区域划分成若干子区域,并依据Fisher准则和AP缺失频率,选取分辨能力强且可靠的AP节点作为子区域的训练节点,最后采用DBN模型对各子区域参考点数据进行训练;在线阶段根据接收信号强度判别测试点所属类簇,并利用训练好的DBN模型在线估计测试点位置。实验结果表明,与WKNN、M-WKNN以及PSO-ANN算法相比,改进算法在定位精度和稳定性方面均有所提高。
关键词 室内定位;聚类;AP选择;DBN;指纹库训练
基金项目 国家自然科学基金资助项目(61372058)
本文URL http://www.arocmag.com/article/01-2018-08-058.html
英文标题 Indoor positioning algorithm based on reliable AP selection and deep belief network
作者英文名 Li Xinchun, Guo Xinxin
机构英文名 a.SchoolofElectrics&InformationEngineering,b.SchoolofGraduateStudies,LiaoningTechnicalUniversity,HuludaoLiaoning125105,China
英文摘要 Due to non-line-of-sight propagation and other effects, the accuracy of indoor positioning based on location fingerprint is not high. Aiming at this problem, this paper proposed a novel indoor positioning algorithm, which based on the reliable AP selection and deep belief network (DBN). Firstly, the algorithm used the improved K-means clustering algorithm to divide the locating area into several sub-regions in the off-line phase. Then according to the Fisher criterion and the AP absent frequency, it selected the strongly distinguishable and reliable AP node as the training node of the sub-region. And finally it used the DBN model to train the data of each sub-region. In the on-line phase, the improved algorithm determined the cluster according to the received signal strength, and estimated the location of test point by the trained DBN model. The experimental results show that, compared with WKNN, M-WKNN and PSO-ANN algorithm, the proposed algorithm can effectively improve the accuracy and stability of positioning.
英文关键词 indoor positioning; clustering; AP selection; DBN; fingerprint database training
参考文献 查看稿件参考文献
  [1] 陈丽娜. WLAN位置指纹室内定位关键技术研究[D] . 上海:华东师范大学, 2014.
[2] Xue Weixing, Qiu Weining, Hua Xianghong, et al. Improved Wi-Fi RSSI measurement for indoor localization[J] . IEEE Sensors Journal, 2017, 17(7):2224-2230.
[3] 王培重, 郑南山, 张言哲. 基于动态K值及AP MAC地址筛选的室内定位算法[J] . 计算机科学, 2016, 43(1):163-165.
[4] 聂尔豪, 于重重, 苏维均, 等. Wi-Fi实时定位算法研究[J] . 计算机应用研究, 2014, 31(7):2164-2167.
[5] 周牧, 蒲巧林, 田增山. 室内WLAN定位中位置指纹优化的接入点部署方法[J] . 通信学报, 2015, 36(S1):30-41.
[6] He Suining, Gary C S H. Wi-Fi fingerprint-based indoor positioning:recent advances and comparisons[J] . IEEE Communications Surveys & Tutorials, 2016, 18(1):466-490.
[7] 张勇, 黄杰, 徐科宇. 基于PCA-LSSVR算法的WLAN室内定位方法[J] . 仪器仪表学报, 2015, 36(2):408-414.
[8] Saha A, Sadhukhan P. A novel clustering strategy for fingerprinting-based localization system to reduce the searching time[C] //Proc of the 2nd International Conference on Recent Trends in Information Systems. Piscataway, NJ:IEEE Press, 2015:538-543.
[9] Jhuang F M, Hung C F, Tuan C C, et al. An AP selection with RSS standard deviation for indoor positioning in Wi-Fi[C] //Proc of the 9th International Conference on Innovative Mobile and Internet Services in Ubiquitous Computing. Washington DC:IEEE Computer Science, 2015:403-407.
[10] Zhou Mu, Pu Qiaolin, Xu Kunjie, et al. Location fingerprint discrimi-nation maximization for indoor WLAN access point optimization using fast discrete water-filling[C] //Proc of Global Communications Conference. Piscataway, NJ:IEEE Press, 2015:1-6.
[11] 郭建, 刘全景, 赵志凯. 一种基于SVR的定位误差修正算法[J] . 计算机应用研究, 2017, 33(8):1-4.
[12] Liu Wen, Fu Xiao, Deng Zhongliang, et al. Smallest enclosing circle-based fingerprint clustering and modified-WKNN matching algorithm for indoor positioning[C] //Proc of International Conference on Indoor Positioning and Indoor Navigation. Piscataway, NJ:IEEE Press, 2016:1-6.
[13] Gharghan S K, Nordin R, Ismail M, et al. Accurate wireless sensor localization technique based on hybrid PSO-ANN algorithm for indoor and outdoor track cycling[J] . IEEE Sensors Journal, 2016, 16(2):529-541.
[14] 李华亮, 钱志鸿, 田洪亮. 基于核函数特征提取的室内定位算法研究[J] . 通信学报, 2017, 38(1):158-167.
[15] Ko C H, Wu S H. A proactive indoor positioning system in randomly deployed dense WiFi networks[C] //Proc of Global Communications Conference. Piscataway, NJ:IEEE Press, 2017:1-6.
[16] 毛红文. 基于模糊聚类的位置指纹室内定位优化技术研究[D] . 昆明:云南大学, 2014.
[17] MacQueen J. Some methods for classification and analysis of multivariate observations[C] //Proc of the 5th Berkeley Symposium on Math, Statistics, and Probability. Berkeley, CA:University of California Press, 1967:281-297.
[18] 葛柳飞. 基于多层神经网络的室内定位算法研究[D] . 徐州:中国矿业大学, 2016.
[19] Cao Suqun, Zhu Quanyin, Li Bokui. Fuzzy Fisher criterion based edge detection[J] . International Journal of Digital Content Technology and its Applications, 2011, 5(8):381-388.
[20] 张兴. WLAN室内定位信号特征提取算法研究[D] . 哈尔滨:哈尔滨工业大学, 2013.
[21] Guo Jianhui, Liu Xianzhong, Wang Zhongzhu. Optimized indoor positioning based on Wi-Fi in mobile classroom project[C] //Proc of the 11th International Conference on Natural Computation. Piscataway, NJ:IEEE Press, 2014:1208-1212.
[22] 郭丽丽, 丁世飞. 深度学习研究进展[J] . 计算机科学, 2015, 42(5):28-33.
[23] 曾安, 郑齐弥. 基于MIC的深度置信网络研究[J] . 计算机科学, 2016, 43(8):249-253.
[24] Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J] . Science, 2006, 313(5786):504-507.
[25] 焦李成, 杨淑媛, 刘芳, 等. 神经网络七十年:回顾与展望[J] . 计算机学报, 2016, 39(8):1697-1716.
[26] Chen Yushi, Zhao Xing, Jia Xiuping. Spectral-spatial classification of hyperspectral data based on deep belief network[J] . IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(6):2381-2392.
收稿日期 2017/4/17
修回日期 2017/5/24
页码 2469-2473
中图分类号 TP393.09
文献标志码 A