基于距离测量和位置指纹的室内定位方法研究
|
摘要
随着WiFi网络在世界范围内的迅速普及和广泛部署,基于WiFi的室内定位技术由于成本低、易于实现受到了广泛关注.其中,基于WiFi的被动式指纹室内定位由于可直接利用现有的商业WiFi设备,且不需要待定位目标携带任何设备,因此部署成本低、易扩展,同时还具有良好的非侵入性,故已逐渐成为室内定位技术研究者们广泛关注的热点.目前,已有基于WiFi的被动式指纹室内定位技术,如Nuzzer和Pilot等,其定位过程一般可分为离线和在线两个阶段.离线阶段采集相应的信号,并存储所有参考点生成的指纹数据以构建离线指纹数据库.在线阶段则通过采用与离线阶段相同的处理方式得到待测位置的在线实测指纹,然后将其与离线指纹库中的已有指纹进行匹配,从而估计目标位置.然而,现有工作由于存在两个重要问题,导致定位的实时性和精度都不能令人满意.其一,现有指纹匹配定位方式由于需要在在线阶段将实测指纹与指纹库中的所有指纹进行一一匹配,所以计算量大从而导致定位过程的实时性较差;其二,由于离线指纹库中存在与目标当前位置相隔较远但相似的指纹,而这些指纹很可能会对指纹匹配过程造成干扰,进而导致定位误差较大.针对上述两个问题,该文结合位置指纹定位技术和距离测量算法,提出了一种新的位置指纹室内定位方法ILLFRM.该方法在在线阶段中加入了粗定位,并在进行指纹匹配之前,通过粗定位来过滤离线指纹库中与目标当前位置不相关的指纹,以减少匹配过程中的计算量和避免不相干指纹的干扰,从而同时达到改善定位精度和实时性的目的.通过在空旷大厅和走廊的真实场景进行实验,结果表明,该文提出的方法与Pilot和Nuzzer相比,定位精度分别提高了约28%和51%.此外,由于一次匹配过程的总耗时不足200ms,因此ILLFRM可以很好地满足实时性要求.
With the rapid development of WiFi technology,the WiFi network has been in existence widely today around the world.Various indoor positioning technologies based on WiFi have constantly emerged,and already aroused wide attention due to their virtues of low cost and easy implementation.Among of them,the WiFi-based passive fingerprint indoor positioning has become a key interest of research since it is cheap,non-invasive,and easy to extend.In other words,it can be directly deployed on the existing business WiFi equipment without any extra devices binding on the objective.Currently,the existing works on passive fingerprint positioning,such as Nuzzer and Pilot,generally contain two phases,offline and online.The offline phase is mainly responsible for collecting the signals corresponding to the objective,and storing the fingerprint data of all the reference points,thus constructing the offline fingerprint database.In the online phase,the measured fingerprints of the target entity are obtained by the same way,and matched with all the fingerprints in the whole offline fingerprint database,thereby estimating the target position.Nonetheless,the timeliness and accuracy of the existing works are not satisfying because they ignore two important facts including:(1)in the online phase performing the match actions throughout the whole offline fingerprint database spends too much time during the localization,especially when fingerprint database is huge;(2)the offline fingerprint database generally involves some fingerprints which are actually far from the current position of the target,but may interfere with the fingerprint matching.These fingerprints are easy to increase the localization errors,which in turn causes the inaccurately positioning.To address the above two problems,this paper proposes ILLFRM(Indoor Localization Method Based on Location Fingerprint and Range Measurement),which is a novel passive fingerprint indoor positioning method that combines the location fingerprint technology with range measurement algorithm.ILLFRM proactively introduces a heuristic operation,namely"coarse positioning",in the localization process.Before the fingerprint matching in the online phase,the coarse positioning in advance filters out those fingerprints irrelevant to the current location of the target in the offline fingerprint database,hence the computation load can be significantly reduced,and the interference from those irrelevant fingerprints in the current offline fingerprint database is greatly eliminated at the same time.By this way,ILLFRM ensures not only the accuracy of the positioning results,but the timeliness of the localization process as well.We have run a series of real implementations of ILLFRM in two typical scenarios.The first one is an empty hall whose area is approximately 80 square meters,while the other one is a corridor with the area of 84 square meters.The former scenario can be seen as the typical open environment,and the latter one can be considered as the representative multi-wall environment.The test results show that ILLFRM has better performance compared to the existing passive fingerprint positioning methods.To be specific,the performance revenue of positioning accuracy is about 28% and 51% compared to Pilot and Nuzzer.Besides,the time duration of match process in ILLFRM is less than 200 milliseconds,which testifies that the timeliness of ILLFRM is satisfying.
With the rapid development of WiFi technology,the WiFi network has been in existence widely today around the world.Various indoor positioning technologies based on WiFi have constantly emerged,and already aroused wide attention due to their virtues of low cost and easy implementation.Among of them,the WiFi-based passive fingerprint indoor positioning has become a key interest of research since it is cheap,non-invasive,and easy to extend.In other words,it can be directly deployed on the existing business WiFi equipment without any extra devices binding on the objective.Currently,the existing works on passive fingerprint positioning,such as Nuzzer and Pilot,generally contain two phases,offline and online.The offline phase is mainly responsible for collecting the signals corresponding to the objective,and storing the fingerprint data of all the reference points,thus constructing the offline fingerprint database.In the online phase,the measured fingerprints of the target entity are obtained by the same way,and matched with all the fingerprints in the whole offline fingerprint database,thereby estimating the target position.Nonetheless,the timeliness and accuracy of the existing works are not satisfying because they ignore two important facts including:(1)in the online phase performing the match actions throughout the whole offline fingerprint database spends too much time during the localization,especially when fingerprint database is huge;(2)the offline fingerprint database generally involves some fingerprints which are actually far from the current position of the target,but may interfere with the fingerprint matching.These fingerprints are easy to increase the localization errors,which in turn causes the inaccurately positioning.To address the above two problems,this paper proposes ILLFRM(Indoor Localization Method Based on Location Fingerprint and Range Measurement),which is a novel passive fingerprint indoor positioning method that combines the location fingerprint technology with range measurement algorithm.ILLFRM proactively introduces a heuristic operation,namely"coarse positioning",in the localization process.Before the fingerprint matching in the online phase,the coarse positioning in advance filters out those fingerprints irrelevant to the current location of the target in the offline fingerprint database,hence the computation load can be significantly reduced,and the interference from those irrelevant fingerprints in the current offline fingerprint database is greatly eliminated at the same time.By this way,ILLFRM ensures not only the accuracy of the positioning results,but the timeliness of the localization process as well.We have run a series of real implementations of ILLFRM in two typical scenarios.The first one is an empty hall whose area is approximately 80 square meters,while the other one is a corridor with the area of 84 square meters.The former scenario can be seen as the typical open environment,and the latter one can be considered as the representative multi-wall environment.The test results show that ILLFRM has better performance compared to the existing passive fingerprint positioning methods.To be specific,the performance revenue of positioning accuracy is about 28% and 51% compared to Pilot and Nuzzer.Besides,the time duration of match process in ILLFRM is less than 200 milliseconds,which testifies that the timeliness of ILLFRM is satisfying.
引文
[1]Youssef M,Agrawala A.The Horus WLAN location determination system//Proceedings of the 3rd International Conference on Mobile Systems,Applications,and Services.Seattle,USA,2005:205-218
[2]Wu C,Yang Z,Liu Y,et al.WILL:Wireless indoor localization without site survey.IEEE Transactions on Parallel and Distributed Systems,2013,24(4):839-848
[3]Seifeldin M,Saeed A,Kosba A E,et al.Nuzzer:A largescale device-free passive localization system for wireless environments.IEEE Transactions on Mobile Computing,2013,12(7):1321-1334
[4]Xiao J,Wu K,Yi Y,et al.Pilot:Passive device-free indoor localization using channel state information//Proceedings of the33rd IEEE International Conference on Distributed Computing Systems(ICDCS 2013).Philadelphia,USA.2013:236-245
[5]Wu K,Xiao J,Yi Y,et al.FiLA:Fine-grained indoor localization//Proceedings of the 31st Annual IEEE International Conference on Computer Communications.Orlando,USA,2012:2210-2218
[6]Chen Ci,Mu Nan,Zhang Chen,et al.Indoor fingerprint positioning model based on principal component analysis.Journal of Software,2013,24(Supplement(1)):98-107(in Chinese)(陈祠,牟楠,张晨等.基于主成分分析的室内指纹定位模型.软件学报,2013,24(增刊(1)):98-107)
[7]Shin B,Lee J H,Lee T,et al.Enhanced weighted K-nearest neighbor algorithm for indoor Wi-Fi positioning systems//Proceedings of the 2012 8th International Conference on Computing Technology and Information Management(ICCM).Seoul,South Korea,2012:574-577
[8]Wang W,Liu A X,Shahzad M,et al.Understanding and modeling of WiFi signal based human activity recognition//Proceedings of the 21st Annual International Conference on Mobile Computing and Networking.Paris,France,2015:65-76
[9]Youssef M,Mah M,Agrawala A.Challenges:Device-free passive localization for wireless environments//Proceedings of the 13th Annual ACM International Conference on Mobile Computing and Networking.Montreal,Canada,2007:222-229
[10]Wu C,Yang Z,Zhou Z,et al.Non-invasive detection of moving and stationary human with WiFi.IEEE Journal on Selected Areas in Communications,2015,33(11):2329-2342
[11]Wu Z,Xu Q,Li J,et al.Passive indoor localization based on CSI and Naive Bayes classification.IEEE Transactions on Systems,Man,and Cybernetics:Systems,2017,48(9):1566-1577
[12]Wang X,Gao L,Mao S.BiLoc:Bi-modal deep learning for indoor localization with commodity 5GHz WiFi.IEEEAccess,2017,5:4209-4220
[13]Wang X,Gao L,Mao S,et al.CSI-based fingerprinting for indoor localization:A deep learning approach.IEEE Transactions on Vehicular Technology,2017,66(1):763-776
[14]Chapre Y,Ignjatovic A,Seneviratne A,et al.CSI-MIMO:Indoor Wi-Fi fingerprinting system//Proceedings of the 39th IEEE Conference on Local Computer Networks(LCN).Edmonton,Canada,2014:202-209
[2]Wu C,Yang Z,Liu Y,et al.WILL:Wireless indoor localization without site survey.IEEE Transactions on Parallel and Distributed Systems,2013,24(4):839-848
[3]Seifeldin M,Saeed A,Kosba A E,et al.Nuzzer:A largescale device-free passive localization system for wireless environments.IEEE Transactions on Mobile Computing,2013,12(7):1321-1334
[4]Xiao J,Wu K,Yi Y,et al.Pilot:Passive device-free indoor localization using channel state information//Proceedings of the33rd IEEE International Conference on Distributed Computing Systems(ICDCS 2013).Philadelphia,USA.2013:236-245
[5]Wu K,Xiao J,Yi Y,et al.FiLA:Fine-grained indoor localization//Proceedings of the 31st Annual IEEE International Conference on Computer Communications.Orlando,USA,2012:2210-2218
[6]Chen Ci,Mu Nan,Zhang Chen,et al.Indoor fingerprint positioning model based on principal component analysis.Journal of Software,2013,24(Supplement(1)):98-107(in Chinese)(陈祠,牟楠,张晨等.基于主成分分析的室内指纹定位模型.软件学报,2013,24(增刊(1)):98-107)
[7]Shin B,Lee J H,Lee T,et al.Enhanced weighted K-nearest neighbor algorithm for indoor Wi-Fi positioning systems//Proceedings of the 2012 8th International Conference on Computing Technology and Information Management(ICCM).Seoul,South Korea,2012:574-577
[8]Wang W,Liu A X,Shahzad M,et al.Understanding and modeling of WiFi signal based human activity recognition//Proceedings of the 21st Annual International Conference on Mobile Computing and Networking.Paris,France,2015:65-76
[9]Youssef M,Mah M,Agrawala A.Challenges:Device-free passive localization for wireless environments//Proceedings of the 13th Annual ACM International Conference on Mobile Computing and Networking.Montreal,Canada,2007:222-229
[10]Wu C,Yang Z,Zhou Z,et al.Non-invasive detection of moving and stationary human with WiFi.IEEE Journal on Selected Areas in Communications,2015,33(11):2329-2342
[11]Wu Z,Xu Q,Li J,et al.Passive indoor localization based on CSI and Naive Bayes classification.IEEE Transactions on Systems,Man,and Cybernetics:Systems,2017,48(9):1566-1577
[12]Wang X,Gao L,Mao S.BiLoc:Bi-modal deep learning for indoor localization with commodity 5GHz WiFi.IEEEAccess,2017,5:4209-4220
[13]Wang X,Gao L,Mao S,et al.CSI-based fingerprinting for indoor localization:A deep learning approach.IEEE Transactions on Vehicular Technology,2017,66(1):763-776
[14]Chapre Y,Ignjatovic A,Seneviratne A,et al.CSI-MIMO:Indoor Wi-Fi fingerprinting system//Proceedings of the 39th IEEE Conference on Local Computer Networks(LCN).Edmonton,Canada,2014:202-209