基于高斯过程的无线信号指纹定位算法
|
摘要
基于Wi-Fi接收信号强度的室内定位技术因其良好的定位性能获得广泛关注。为了提高定位精度,需要构建高密度信号指纹数据库。但是创建与维护数据库需要投入大量人力物力,高密度信号指纹数据库还会增加定位所需时间,实时性难以得到保证。针对上述问题,提出一种基于高斯过程的两阶段信号指纹定位算法:利用高斯过程,在训练数据库基础上,生成高密度虚拟信号指纹数据库;定位过程中首先使用训练数据库确定大致范围,然后使用虚拟数据库进行精确定位。仿真结果显示,该算法与K近邻算法相比定位精度平均提高了81.9%,定位时间平均减少86%,从而验证了算法的有效性。
Indoor localization based on Wi-Fi received signal strength(RSS) has attracted wide attention because of its good positioning performance. In order to improve the localization accuracy, it is necessary to build a high-density fingerprint database. However, the creation and maintenance of database requires a lot of manpower and material resources. High-density fingerprint database increases the time required for location, and it is difficult to guarantee real-time. For the above problems, we proposed a two-stage signal fingerprint localization algorithm based on Gaussian process. Based on the training database, a high-density virtual signal fingerprint database was generated by using the Gaussian process. In the localization process, the training database was used to determine the approximate range, and we used the virtual database to locate accurately. The simulation results show that the algorithm improves the localization accuracy by 81.9% and reduces the localization time by 86% on average compared with K-nearest neighbor algorithm, thus verifying its effectiveness.
Indoor localization based on Wi-Fi received signal strength(RSS) has attracted wide attention because of its good positioning performance. In order to improve the localization accuracy, it is necessary to build a high-density fingerprint database. However, the creation and maintenance of database requires a lot of manpower and material resources. High-density fingerprint database increases the time required for location, and it is difficult to guarantee real-time. For the above problems, we proposed a two-stage signal fingerprint localization algorithm based on Gaussian process. Based on the training database, a high-density virtual signal fingerprint database was generated by using the Gaussian process. In the localization process, the training database was used to determine the approximate range, and we used the virtual database to locate accurately. The simulation results show that the algorithm improves the localization accuracy by 81.9% and reduces the localization time by 86% on average compared with K-nearest neighbor algorithm, thus verifying its effectiveness.
引文
[1]郝娟,张著洪,凃歆.噪声环境下RSSI定位问题及其求解的果蝇优化算法[J].计算机工程与应用,2018,54(7):121-126.
[2]Górak R,Luckner M,Okulewicz M,et al.Indoor Localisation Based on GSM Signals:Multistorey Building Study[J].Mobile Information Systems,2016,2016(Article ID 2719576):1-17.
[3]Yoon S,Lee K,Yun Y C,et al.ACMI:FM-Based Indoor Localization via Autonomous Fingerprinting[J].IEEE Transactions on Mobile Computing,2016,15(6):1318-1332.
[4]Wu H,Wang Q,Zhao Y,et al.Indoor localization using FM radio and DTMB signals[J].Radio Science,2016,51(7):1030-1037.
[5]Lawrence B F D F N.WiFi-SLAM Using Gaussian Process Latent Variable Models[C]//International Joint Conference on Artifical Intelligence.Morgan Kaufmann Publishers Inc.2007:2480-2485.
[6]Han D,Han D,Moon B,et al.Address-Based Crowdsourcing Radio Map Construction for Wi-Fi Positioning Systems[C]//International Conference on Indoor Positioning&Indoor Navigation.IEEE,2015.
[7]Wang B,Chen Q,Yang L T,et al.Indoor smartphone localization via fingerprint crowdsourcing:challenges and approaches[J].IEEE Wireless Communications,2016,23(3):82-89.
[8]Zhou X,Chen T,Guo D,et al.From one to crowd:a survey on crowdsourcing-based wireless indoor localization[J].Frontiers of Computer Science,2018(1):1-28.
[9]Rappaport T S.Wireless communications:principles and practice[M].Prentice Hall PTR,New Jersey,1996.
[10]周牧,唐云霞,田增山,等.基于流形插值数据库构建的WLAN室内定位算法[J].电子与信息学报,2017,39(8):1826-1834.
[11]田增山,代海鹏.基于曲面拟合的WiFi指纹数据库更新[J].计算机应用,2016,36(5):1192-1195.
[12]Wang H,Deng Z,Fu X,et al.Ray-Tracing Aided Indoor Bayesian Positioning Algorithm[C]//China Satellite Navigation Conference(CSNC)2018 Proceedings.2018.
[13]Zhang M,Pei L,Deng X.Graph SLAM-based Crowdsourcing framework for indoor Wi-Fi fingerprinting[C]//International Conference on Ubiquitous Positioning.IEEE,2017:61-67.
[14]Clark J.Hierarchical geometric models for visible surface algorithms[J].Communications of The ACM,1976,19(10):547-554.
[15]Ferris B,Hahnel D,Fox D.Gaussian Processes for Signal Strength-Based Location Estimation[C]//Robotics:Science and Systems II.University of Pennsylvania,Philadelphia,Pennsylvania,USA,2006.
[2]Górak R,Luckner M,Okulewicz M,et al.Indoor Localisation Based on GSM Signals:Multistorey Building Study[J].Mobile Information Systems,2016,2016(Article ID 2719576):1-17.
[3]Yoon S,Lee K,Yun Y C,et al.ACMI:FM-Based Indoor Localization via Autonomous Fingerprinting[J].IEEE Transactions on Mobile Computing,2016,15(6):1318-1332.
[4]Wu H,Wang Q,Zhao Y,et al.Indoor localization using FM radio and DTMB signals[J].Radio Science,2016,51(7):1030-1037.
[5]Lawrence B F D F N.WiFi-SLAM Using Gaussian Process Latent Variable Models[C]//International Joint Conference on Artifical Intelligence.Morgan Kaufmann Publishers Inc.2007:2480-2485.
[6]Han D,Han D,Moon B,et al.Address-Based Crowdsourcing Radio Map Construction for Wi-Fi Positioning Systems[C]//International Conference on Indoor Positioning&Indoor Navigation.IEEE,2015.
[7]Wang B,Chen Q,Yang L T,et al.Indoor smartphone localization via fingerprint crowdsourcing:challenges and approaches[J].IEEE Wireless Communications,2016,23(3):82-89.
[8]Zhou X,Chen T,Guo D,et al.From one to crowd:a survey on crowdsourcing-based wireless indoor localization[J].Frontiers of Computer Science,2018(1):1-28.
[9]Rappaport T S.Wireless communications:principles and practice[M].Prentice Hall PTR,New Jersey,1996.
[10]周牧,唐云霞,田增山,等.基于流形插值数据库构建的WLAN室内定位算法[J].电子与信息学报,2017,39(8):1826-1834.
[11]田增山,代海鹏.基于曲面拟合的WiFi指纹数据库更新[J].计算机应用,2016,36(5):1192-1195.
[12]Wang H,Deng Z,Fu X,et al.Ray-Tracing Aided Indoor Bayesian Positioning Algorithm[C]//China Satellite Navigation Conference(CSNC)2018 Proceedings.2018.
[13]Zhang M,Pei L,Deng X.Graph SLAM-based Crowdsourcing framework for indoor Wi-Fi fingerprinting[C]//International Conference on Ubiquitous Positioning.IEEE,2017:61-67.
[14]Clark J.Hierarchical geometric models for visible surface algorithms[J].Communications of The ACM,1976,19(10):547-554.
[15]Ferris B,Hahnel D,Fox D.Gaussian Processes for Signal Strength-Based Location Estimation[C]//Robotics:Science and Systems II.University of Pennsylvania,Philadelphia,Pennsylvania,USA,2006.