基于射线跟踪的特征匹配定位技术研究
|
摘要
近年来,伴随着移动通信系统快速发展,室内无线定位技术逐渐成为研究的焦点。室内定位技术主要应用于车站、医院、机场、商场、博物馆等环境,其服务可以帮助应对引导、跟踪、消防、反恐等紧急事件。
在室内主要运用基于基站和移动终端的定位技术时,由于建筑理念、风格成千上万,建筑材料、结构、尺度、摆设等各不相同,其带来的非视距传播严重影响了定位精度,成为限制这类定位技术应用于室内的关键因素。基于信号地图的定位技术是克服非视距的主要方法之一。
在研究了基于射线跟踪的序列定位算法后,提出了基于小波消噪的射线跟踪序列匹配定位算法。该算法通过小波变换阈值消噪方法对目标的冲激响应序列进行消噪处理,然后用消噪后的序列与已知位置信标点的冲激响应序列进行匹配定位。仿真表明算法具有较高的定位成功率,有效地降低了环境噪声对定位特征的影响。
在对比了目前基于信号地图的城市及室内定位技术的基础上,提出基于小波变换的特征匹配射线跟踪定位算法。该算法利用射线跟踪算法从接收点反向跟踪所有多径信号并以多径信号的冲激响应来模拟信道,与实测信道建模方式相比简化了工作量。然后通过小波变换从冲激响应序列中分解出包含定位特征的小波系数矩阵,该系数矩阵不仅能够反映信标点发出无线电波在传播过程中信号强度的衰减情况,而且包含信号强度衰减的时间信息,比用信号强度作为定位特征提高了完备性。为了方便建立信号地图,减少匹配算法的复杂度,将小波系数矩阵中分解得到的奇异值向量作为建立本算法信号地图的特征。计算机仿真表明,本算法提取的定位特征具有良好的稳定性,相比用接收信号强度作为特征的定位算法,本算法可以得到更精确的定位结果。算法的定位精度符合室内定位服务要求,而且随着区域内信标点密度增大而提高。
With the fast development of mobile communication system in recent years, the indoor wireless location technology is gradually becoming a research highlight. It is applied in the various circumstances including station, hospital, airport, mall, museums and so on, and its outcome is used for handling with emergency such as leading, guidance, fire prevention and counter-terrorism.
The location technology based on base stations and mobile equipments cannot work in the indoor environment because of the Non-Line-of-sight (NLOS) propagation. The different architectural concept, style, structure, size or display brings out the NLOS propagation which is the key factor to limit this kind of location technology. The location technology based on signal map is one of major ways to cope with the NLOS propagation.
After a number of researches on the sequence location algorithm based on ray-tracing, the sequence of ray-tracing location algorithm based on wavelet de-noising is presented. The de-noising threshold based on wavelet transform is utilized for eliminating the noise from the impulse response of the target, and then the impulse response compares with the ones of position-known beacons which are pre-built by the ray-tracing technology to estimate user location. The simulation demonstrates that the algorithm has a high success rates to locate users’positions, and reduces the impact of the noise effectively.
After the analysis of the existing city and indoor location technologies based on signal map, this paper further puts forward a ray-tracing location algorithm related with channel feature extraction and matching based on wavelet transform. This algorithm first obtains the multi-path signals from the receiver reversely and calculates their impulse response using the ray-tracing method, which simplifies the work comparing with the realistic measurements modeling. Then we extract the feature of the channel from the wavelet transform coefficient matrix of the discrete impulse response sequence. The feature of the channel not only contains the information of how much signal strength decline, but also when the decline happens. It is more complete than the signal strength sequence. In order to build the signal map conveniently and reduce the complication of the matching algorithm, the singular value vector extracted from the wavelet coefficient matrix by singular value decomposition is used as the feature of the signal site. Computer simulation shows that the character of location calculated from the algorithm has excellent stabilization. The location results demonstrate the ability of our algorithm to estimate user location with a higher degree of the accuracy than the algorithm that uses the receive signal strength as the characteristic of the signal map. The overall accuracy of locating objects satisfies the need of the inside location service, and the performance becomes better as the beacon density increases.
在室内主要运用基于基站和移动终端的定位技术时,由于建筑理念、风格成千上万,建筑材料、结构、尺度、摆设等各不相同,其带来的非视距传播严重影响了定位精度,成为限制这类定位技术应用于室内的关键因素。基于信号地图的定位技术是克服非视距的主要方法之一。
在研究了基于射线跟踪的序列定位算法后,提出了基于小波消噪的射线跟踪序列匹配定位算法。该算法通过小波变换阈值消噪方法对目标的冲激响应序列进行消噪处理,然后用消噪后的序列与已知位置信标点的冲激响应序列进行匹配定位。仿真表明算法具有较高的定位成功率,有效地降低了环境噪声对定位特征的影响。
在对比了目前基于信号地图的城市及室内定位技术的基础上,提出基于小波变换的特征匹配射线跟踪定位算法。该算法利用射线跟踪算法从接收点反向跟踪所有多径信号并以多径信号的冲激响应来模拟信道,与实测信道建模方式相比简化了工作量。然后通过小波变换从冲激响应序列中分解出包含定位特征的小波系数矩阵,该系数矩阵不仅能够反映信标点发出无线电波在传播过程中信号强度的衰减情况,而且包含信号强度衰减的时间信息,比用信号强度作为定位特征提高了完备性。为了方便建立信号地图,减少匹配算法的复杂度,将小波系数矩阵中分解得到的奇异值向量作为建立本算法信号地图的特征。计算机仿真表明,本算法提取的定位特征具有良好的稳定性,相比用接收信号强度作为特征的定位算法,本算法可以得到更精确的定位结果。算法的定位精度符合室内定位服务要求,而且随着区域内信标点密度增大而提高。
With the fast development of mobile communication system in recent years, the indoor wireless location technology is gradually becoming a research highlight. It is applied in the various circumstances including station, hospital, airport, mall, museums and so on, and its outcome is used for handling with emergency such as leading, guidance, fire prevention and counter-terrorism.
The location technology based on base stations and mobile equipments cannot work in the indoor environment because of the Non-Line-of-sight (NLOS) propagation. The different architectural concept, style, structure, size or display brings out the NLOS propagation which is the key factor to limit this kind of location technology. The location technology based on signal map is one of major ways to cope with the NLOS propagation.
After a number of researches on the sequence location algorithm based on ray-tracing, the sequence of ray-tracing location algorithm based on wavelet de-noising is presented. The de-noising threshold based on wavelet transform is utilized for eliminating the noise from the impulse response of the target, and then the impulse response compares with the ones of position-known beacons which are pre-built by the ray-tracing technology to estimate user location. The simulation demonstrates that the algorithm has a high success rates to locate users’positions, and reduces the impact of the noise effectively.
After the analysis of the existing city and indoor location technologies based on signal map, this paper further puts forward a ray-tracing location algorithm related with channel feature extraction and matching based on wavelet transform. This algorithm first obtains the multi-path signals from the receiver reversely and calculates their impulse response using the ray-tracing method, which simplifies the work comparing with the realistic measurements modeling. Then we extract the feature of the channel from the wavelet transform coefficient matrix of the discrete impulse response sequence. The feature of the channel not only contains the information of how much signal strength decline, but also when the decline happens. It is more complete than the signal strength sequence. In order to build the signal map conveniently and reduce the complication of the matching algorithm, the singular value vector extracted from the wavelet coefficient matrix by singular value decomposition is used as the feature of the signal site. Computer simulation shows that the character of location calculated from the algorithm has excellent stabilization. The location results demonstrate the ability of our algorithm to estimate user location with a higher degree of the accuracy than the algorithm that uses the receive signal strength as the characteristic of the signal map. The overall accuracy of locating objects satisfies the need of the inside location service, and the performance becomes better as the beacon density increases.
引文
[1] D. Madigan, E Elnahrawy. GPS Landing System Reference Antenna [J]. IEEE Antennas & Propagation,2010,52(2):1567-1573
[2]高路.双星定位的导航系统[J].现代通信,1997:20
[3]范平志,邓平等.蜂窝网无线定位[M].北京:电子工业出版社,2002.12:22-34.
[4]孙巍,王行刚.移动定位系统综述[J].电子技术应用,2003:20
[5]顾杰.无线蜂窝移动通信系统中移动台定位技术研究与实现[D].成都:电子科技大学,2003
[6]杨天池,金梁等.一种基于TOA定位的CHAN算法改进算法[J].电子学报,2009.4:819-822
[7]王欣,刘宗兴等.一种考虑非视线传播影响的TOA定位算法[J].通信学报,2001,3:2-7
[8]郭福成,李宗华,孙仲康.无线定位跟踪修正协方差扩展卡尔曼滤波算法[J].电子与信息学报,2004:26(6):83-90
[9] C, Kassis, J Picheral, C Mokbel .Advantages of nonuniform arrays using root-MUSIC[J] Signal Processing, .2010,90(2)
[10] Zhigang Liu, Jinkuan Wang, Fuli Wang .Conjugate unitary ESPRIT for real sources adapted to the coherent case[J] Signal Processing, .2009,89(7)
[11] Bahl Paramvir, Venkata N Padmanabhan. RADAR: An In-Building RF-based User Location and Tracking System[J] Proc IEEE Info Com, pp.2000:775-784
[12] L. Ni, Y.Liu. LANDMARC: Indoor Loctation Sensing Using Active RFID[C]. Proc of the First IEEE International Conference on Pervasive Computing and Communication 2003
[13] K. Lorincz, M.welsh. MoteTrack: A Robust, Decentralized approach to RF-Based Location Tracing[C].Proc of the International Workshop on Location and Context-Awarenes, LNCS 2005:63-82
[14]季忠.移动通信的电波预测和基站优化研究[D].上海:上海交通大学,2000
[15]谢勇.电子地图开发及微蜂窝区域电波预测算法优化[D].上海:上海交通大学,2003
[16]袁正午.蜂窝通信系统移动终端射线跟踪定位理论与方法研究[D].湖南,中南大学,2003
[17]吴志忠.移动通信无线电波传播[M].北京:人民邮电出版社,2002.9:22-34.
[18]郭辉平,刘学观.电磁场与电磁波[M].西安:西安电子科技大学出版社,2004.12:5-10.
[19]张瑜.电磁波空间传播[M]西安:西安电子科技大学出版社,2007.12:4-8.
[20] G.L.Turin, T.L.Johnson, S.B.Fine. A Statistical Model of Urban Multipath Propagation[J]. IEEE Trans. Vehic Technol.1972,21:1-9.
[21] Akihiro kajiwara. Millimeter-wave indoor radio channel with artificial reflector[J]. IEEE Trans. Vehic Technol. 1999,46:486-493.
[22] Emad Torabi,Ayaz Ghorbani.Reflection coefficient modification for mobile path loss calculation[J].Days on Diffration.2005:94-107
[23] J.B.Keller. A symptotic solution of some diffraction problem[J].Comm Pure and Appl Math,1956,(9):207-265.
[24] J.B.Keller. Geometrical theory of diffraction[J].J.Opt.Soc,1962,(52):116-130.
[25] P.D.Holm. A new heuristic UTD diffraction coefficient for nonperfectly conducting wedges[J]. IEEE Trans. AP, Aug,2000,48(8):1211-1219.
[26] K.A.Chamberlin, R.J.Luebbers. An evaluation of Longley-Rice and GTD propagation model[J]. IEEE Trans. AP, Nov, 1982, 30 (6):1093-1098.
[27] R.J.Luebbers. Finite conductivity uniform GTD verse knife edge diffraction in prediction of propagation path loss[J]. IEEE Trans. AP, Jan,1984, 32 (1):70-76.
[28]李超峰.MIMO信道电波传播特性[D].西安:西安电子科技大学,2006
[29]朱若萍.基于射线跟踪模型的微蜂窝小区电波传播预测研究[D].成都:电子科技大学,2007
[30]程勇,吴剑锋,曹伟.一种用于移动系统场强预测的准三维射线跟踪模型[J].电波科学学报,2002
[31]陈建春.矢量图形系统开发与编程[M].北京:电子工业出版社,2004.1:395-396
[32]唐克.MFC程序设计[M].北京:北京希望电子出版社.2002:76-92
[33] S, Lippman,李师贤等译.C++ primer (4th Edition)[M].北京:人民邮电出版社.2006:272-284
[34]张曜,郭立山,李缴玄.Visual C++(MFC)函数实用手册[M].武汉:冶金工业出版社.2004,1:263-264
[35]袁正午.移动终端射线跟踪定位理论与方法[M].北京:电子工业出版社,2007,7:72-94.
[36]汪茂光.几何绕射理论[M].西安:西安电子科技大学出版社.1985:64-83.
[37]胡银彪,昂海松.射线跟踪法的改进及其应用[J].南京航天航空大学学报,2004,8:501-504
[38] R.Valenuela. Ray-tracing Prediction of indoor Radio Propagation[C]. PIMAC94,1994:140-144.
[39] Xiaoyang Huang, J Stames, M Farwell. Radio-Propagation Model Based on the Combined Method of Ray-tracing and Diffraction [J]. IEEE Trans.Antennas & Propagation,2006, 54 (4):1284-1290.
[40]彭玉华.小波变换与工程应用[M].北京:科学出版社.2002:23-34.
[41]杨福生.小波变换的工程分析与应用[M].北京:科学出版社.2003:15-20.
[42]李弼程,罗建书.小波分析与其应用[M].北京:电子工业出版社.2003:43-45.
[43]彭启琮,邵怀宗,李明奇.信号分析[M].北京:电子工业出版社.2006:145-168.
[44] S.Mallat杨力华译.信号处理的小波分析引导[M].北京:机械工业出版社.2002:54-58.
[45]杜浩藩,丛爽.基于Matlab小波去噪方法的研究[J].计算机仿真.2003,7:119-121.
[46]徐长发,李国宽.实用小波方法[M].武汉:华中科技大学出版社.2003:65-70.
[47]朱洪俊,王忠,秦树人.小波变换对瞬态信号特征信息的精确提取[J].机械工程学报,2005,12:196-199.
[48]孙贵清,杨德森.小波变换在瞬态检测中的应用[J].哈尔滨工程大学学报,1998,2:42-46.
[49]钟伯成,吴小培.基于小波变换的脑电信号的瞬态特征提取[J].模式识别与人工智能,2000,6:218-221
[50] Zhu Qibing; Yang Huizhong. Extraction of fault feature in gear system based on convolution type of wavelet packet transform and singular value decomposition [C]. Control Conference, 2008. CCC 2008. 27th Chinese Digital Object Identifier 2008,6:21-24
[51]闵志华,孙利民,孙智等.基于小波变换和奇异值分解的模态参数识别[J].同济大学学报(自然科学版),2009,4:460-465
[52] H.Demirel, C.Ozcinar G.Anbarjafari. Satellite Image Contrast Enhancement Using Discrete Wavelet Transform and Singular Value Decomposition[C]. Geoscience and Remote Sensing Letters, IEEE 2010,4:333-337.
[53]赵学智,叶邦彦,陈统坚.多分辨奇异值理论及其在信号处理和故障诊断中的应用[J].机械工程学报,2010,10:64-74
[54]张贤达.矩阵分析与应用[M].北京:清华大学出版社,2004:102-104.
[2]高路.双星定位的导航系统[J].现代通信,1997:20
[3]范平志,邓平等.蜂窝网无线定位[M].北京:电子工业出版社,2002.12:22-34.
[4]孙巍,王行刚.移动定位系统综述[J].电子技术应用,2003:20
[5]顾杰.无线蜂窝移动通信系统中移动台定位技术研究与实现[D].成都:电子科技大学,2003
[6]杨天池,金梁等.一种基于TOA定位的CHAN算法改进算法[J].电子学报,2009.4:819-822
[7]王欣,刘宗兴等.一种考虑非视线传播影响的TOA定位算法[J].通信学报,2001,3:2-7
[8]郭福成,李宗华,孙仲康.无线定位跟踪修正协方差扩展卡尔曼滤波算法[J].电子与信息学报,2004:26(6):83-90
[9] C, Kassis, J Picheral, C Mokbel .Advantages of nonuniform arrays using root-MUSIC[J] Signal Processing, .2010,90(2)
[10] Zhigang Liu, Jinkuan Wang, Fuli Wang .Conjugate unitary ESPRIT for real sources adapted to the coherent case[J] Signal Processing, .2009,89(7)
[11] Bahl Paramvir, Venkata N Padmanabhan. RADAR: An In-Building RF-based User Location and Tracking System[J] Proc IEEE Info Com, pp.2000:775-784
[12] L. Ni, Y.Liu. LANDMARC: Indoor Loctation Sensing Using Active RFID[C]. Proc of the First IEEE International Conference on Pervasive Computing and Communication 2003
[13] K. Lorincz, M.welsh. MoteTrack: A Robust, Decentralized approach to RF-Based Location Tracing[C].Proc of the International Workshop on Location and Context-Awarenes, LNCS 2005:63-82
[14]季忠.移动通信的电波预测和基站优化研究[D].上海:上海交通大学,2000
[15]谢勇.电子地图开发及微蜂窝区域电波预测算法优化[D].上海:上海交通大学,2003
[16]袁正午.蜂窝通信系统移动终端射线跟踪定位理论与方法研究[D].湖南,中南大学,2003
[17]吴志忠.移动通信无线电波传播[M].北京:人民邮电出版社,2002.9:22-34.
[18]郭辉平,刘学观.电磁场与电磁波[M].西安:西安电子科技大学出版社,2004.12:5-10.
[19]张瑜.电磁波空间传播[M]西安:西安电子科技大学出版社,2007.12:4-8.
[20] G.L.Turin, T.L.Johnson, S.B.Fine. A Statistical Model of Urban Multipath Propagation[J]. IEEE Trans. Vehic Technol.1972,21:1-9.
[21] Akihiro kajiwara. Millimeter-wave indoor radio channel with artificial reflector[J]. IEEE Trans. Vehic Technol. 1999,46:486-493.
[22] Emad Torabi,Ayaz Ghorbani.Reflection coefficient modification for mobile path loss calculation[J].Days on Diffration.2005:94-107
[23] J.B.Keller. A symptotic solution of some diffraction problem[J].Comm Pure and Appl Math,1956,(9):207-265.
[24] J.B.Keller. Geometrical theory of diffraction[J].J.Opt.Soc,1962,(52):116-130.
[25] P.D.Holm. A new heuristic UTD diffraction coefficient for nonperfectly conducting wedges[J]. IEEE Trans. AP, Aug,2000,48(8):1211-1219.
[26] K.A.Chamberlin, R.J.Luebbers. An evaluation of Longley-Rice and GTD propagation model[J]. IEEE Trans. AP, Nov, 1982, 30 (6):1093-1098.
[27] R.J.Luebbers. Finite conductivity uniform GTD verse knife edge diffraction in prediction of propagation path loss[J]. IEEE Trans. AP, Jan,1984, 32 (1):70-76.
[28]李超峰.MIMO信道电波传播特性[D].西安:西安电子科技大学,2006
[29]朱若萍.基于射线跟踪模型的微蜂窝小区电波传播预测研究[D].成都:电子科技大学,2007
[30]程勇,吴剑锋,曹伟.一种用于移动系统场强预测的准三维射线跟踪模型[J].电波科学学报,2002
[31]陈建春.矢量图形系统开发与编程[M].北京:电子工业出版社,2004.1:395-396
[32]唐克.MFC程序设计[M].北京:北京希望电子出版社.2002:76-92
[33] S, Lippman,李师贤等译.C++ primer (4th Edition)[M].北京:人民邮电出版社.2006:272-284
[34]张曜,郭立山,李缴玄.Visual C++(MFC)函数实用手册[M].武汉:冶金工业出版社.2004,1:263-264
[35]袁正午.移动终端射线跟踪定位理论与方法[M].北京:电子工业出版社,2007,7:72-94.
[36]汪茂光.几何绕射理论[M].西安:西安电子科技大学出版社.1985:64-83.
[37]胡银彪,昂海松.射线跟踪法的改进及其应用[J].南京航天航空大学学报,2004,8:501-504
[38] R.Valenuela. Ray-tracing Prediction of indoor Radio Propagation[C]. PIMAC94,1994:140-144.
[39] Xiaoyang Huang, J Stames, M Farwell. Radio-Propagation Model Based on the Combined Method of Ray-tracing and Diffraction [J]. IEEE Trans.Antennas & Propagation,2006, 54 (4):1284-1290.
[40]彭玉华.小波变换与工程应用[M].北京:科学出版社.2002:23-34.
[41]杨福生.小波变换的工程分析与应用[M].北京:科学出版社.2003:15-20.
[42]李弼程,罗建书.小波分析与其应用[M].北京:电子工业出版社.2003:43-45.
[43]彭启琮,邵怀宗,李明奇.信号分析[M].北京:电子工业出版社.2006:145-168.
[44] S.Mallat杨力华译.信号处理的小波分析引导[M].北京:机械工业出版社.2002:54-58.
[45]杜浩藩,丛爽.基于Matlab小波去噪方法的研究[J].计算机仿真.2003,7:119-121.
[46]徐长发,李国宽.实用小波方法[M].武汉:华中科技大学出版社.2003:65-70.
[47]朱洪俊,王忠,秦树人.小波变换对瞬态信号特征信息的精确提取[J].机械工程学报,2005,12:196-199.
[48]孙贵清,杨德森.小波变换在瞬态检测中的应用[J].哈尔滨工程大学学报,1998,2:42-46.
[49]钟伯成,吴小培.基于小波变换的脑电信号的瞬态特征提取[J].模式识别与人工智能,2000,6:218-221
[50] Zhu Qibing; Yang Huizhong. Extraction of fault feature in gear system based on convolution type of wavelet packet transform and singular value decomposition [C]. Control Conference, 2008. CCC 2008. 27th Chinese Digital Object Identifier 2008,6:21-24
[51]闵志华,孙利民,孙智等.基于小波变换和奇异值分解的模态参数识别[J].同济大学学报(自然科学版),2009,4:460-465
[52] H.Demirel, C.Ozcinar G.Anbarjafari. Satellite Image Contrast Enhancement Using Discrete Wavelet Transform and Singular Value Decomposition[C]. Geoscience and Remote Sensing Letters, IEEE 2010,4:333-337.
[53]赵学智,叶邦彦,陈统坚.多分辨奇异值理论及其在信号处理和故障诊断中的应用[J].机械工程学报,2010,10:64-74
[54]张贤达.矩阵分析与应用[M].北京:清华大学出版社,2004:102-104.