Arama Görseller Haritalar Play YouTube Haberler Gmail Drive Daha fazlası »
Oturum açın
Ekran okuyucu kullanıcıları: Erişilebilirlik modu için bu bağlantıyı tıklayın. Erişilebilirlik modu aynı temel özelliklere sahiptir, ancak okuyucunuzla daha iyi çalışır.

Patentler

  1. Gelişmiş Patent Arama
Yayınlanma numarasıCN103079269 A
Yayın türüBaşvuru
Başvuru numarasıCN 201310029536
Yayın tarihi1 May 2013
Dosya kabul tarihi25 Oca 2013
Rüçhan tarihi25 Oca 2013
Yayınlanma numarası201310029536.1, CN 103079269 A, CN 103079269A, CN 201310029536, CN-A-103079269, CN103079269 A, CN103079269A, CN201310029536, CN201310029536.1
Buluş Sahipleri马琳, 周才发, 徐玉滨, 秦丹阳, 孟维晓, 崔扬
Başvuru sahibi哈尔滨工业大学
Alıntıyı Dışa AktarBiBTeX, EndNote, RefMan
Dış Bağlantılar:  SIPO, Espacenet
LDE (Linear Discriminant Analysis) algorithm-based WiFi (Wireless Fidelity) indoor locating method
CN 103079269 A
Özet
The invention discloses an LDE (Linear Discriminant Analysis) algorithm-based WiFi (Wireless Fidelity) indoor locating method, relating to an indoor locating method and solving the problem of poor location instantaneity of the existing WiFi indoor locating method. The realizing process of the LDE algorithm-based WiFi indoor locating method comprises two stages of an offline stage and an online stage, wherein the offline stage comprises the steps of: constructing a WiFi network, measuring RSS (Received Signal Strength) and constructing a Radio Mao; estimating an intrinsic dimension of the Radio Map by adopting an intrinsic dimension estimating method; carrying out dimension reduction process on the Radio Map by adopting an LDE algorithm to obtain a Radio Map subjected to dimension reduction process and a feature transform matrix, wherein an optimal dimension reduction result and a corresponding feature transform matrix are used as a matching database and a corresponding RSS transform matrix in the online stage. The online stage comprises the steps of: carrying out feature transform on the RSS received in a testing point, matching by adopting a KNN (k-Nearest Neighbor) algorithm and the Radio Map subjected to dimension reduction process to obtain predicted coordinates of the testing point. The invention is suitable for indoor location.
Hak Talepleri(7)  şu dilden çevrildi: Çince
1.基于LDE算法的WiFi室内定位方法,其特征是:它由以下步骤实现: 步骤一、针对室内环境布置N个接入点AP,确保所述环境中任意一点被一个或一个以上的接入点AP发出的信号覆盖,所述N个接入点AP组成WiFi网络;在所述室内环境中均匀设置Nkp个参考点;N和Nkp均为正整数; 步骤二、选取一个参考点为坐标原点建立二维直角坐标系,获得Nkp个参考点在该二维直角坐标系中的坐标位置,在离线阶段中在每个参考点上利用信号接收机采集来自每一个接入点AP的信号强度RSS值,并作为该接入点AP的位置特征信息;并根据N个接入点AP的位置特征信息构建室内信号覆盖图Radio Map ; 步骤三、采用本征维数估计算法对步骤二获得的室内信号覆盖图Radio Map进行本征维数分析,获得本征维数分析结果; 步骤四、根据步骤三获得的本征维数分析结果采用LED算法将室内信号覆盖图RadioMap内的所有参考点降维至本征维数,获得特征变换矩阵,并生成降维后的信号覆盖图Radio Map*; 步骤五、在在线阶段,测量室内环境中欲定位点的信号强度RSS值,并将该信号强度RSS值与步骤四获得的特征变换矩阵相乘,获得信号强度变换值RSS* ; 步骤六、采用KNN算法对步骤五获得的信号强度变换值RSS*与步骤四生成的降维后的信号覆盖图Radio Map*进行位置匹配,获得欲定位点的位置坐标,完成欲定位点的室内定位。 1. WiFi-based indoor positioning methods LDE algorithm, characterized in that: it consists of the following steps to achieve: Step 1, for the indoor environment layout of the N AP, and to ensure that the environment at any point by one or more access AP coverage given point, the composition of N WiFi access points AP network; uniform set Nkp reference point in the indoor environment; N and Nkp are positive integers; step two, select a reference point for the coordinate origin the establishment of a two-dimensional Cartesian coordinate system, get Nkp reference point coordinate position in the two-dimensional Cartesian coordinate system, and in the offline stage, using the signal receivers collect from each access point AP signal strength at each reference point RSS value, and as the location of the access point AP feature information; and build indoor signal coverage maps based on the location of Radio Map feature N access points AP information; Step three, using the intrinsic dimension estimation algorithm for indoor obtained in step two Radio Map coverage map were intrinsic dimension analysis to obtain intrinsic dimension analysis; Step four, LED algorithm based on the intrinsic dimension analysis step III was the result of all reference points within the indoor signal coverage map RadioMap dimension reduction to intrinsic dimension, get feature transformation matrix, and generate lower coverage map dimensionality Radio Map *; Step five, in-line stage, I want to measure the signal strength of the indoor environment RSS anchor point value, and the signal strength RSS Step four characteristic values obtained by multiplying the transformation matrix to obtain signal strength conversion value RSS *; step six, using KNN algorithm to convert the signal strength coverage map RSS * value obtained in step Step Five Four generations after dimensionality reduction Radio Map * position match, you want to get the position coordinates of the anchor point to complete the indoor positioning For positioning points.
2.根据权利要求1所述的基于LDE算法的WiFi室内定位方法,其特征在于步骤三中采用本征维数估计算法为特征值估计法。 The WiFi indoor location based LDE algorithm according to claim 1, wherein the step of three adopted intrinsic dimension estimation of eigenvalues estimation.
3.根据权利要求1所述的基于LDE算法的WiFi室内定位方法,其特征在于步骤三中采用本征维数估计算法为包数估计法。 3. The method is based on LDE WiFi indoor positioning algorithm according to claim 1, wherein the step of three adopted intrinsic dimension estimation method to estimate the number of packets.
4.根据权利要求1所述的基于LDE算法的WiFi室内定位方法,其特征在于步骤三中采用本征维数估计算法为测地线最小生成树算法。 4. The WiFi indoor location based LDE algorithm according to claim 1, wherein the step of three adopted intrinsic dimension estimation algorithm geodesic minimum spanning tree algorithm.
5.根据权利要求4所述的基于LDE算法的WiFi室内定位方法,其特征在于采用测地线最小生成树算法对步骤二获得的室内信号覆盖图Radio Map进行本征维数分析是通过公式: 5. The algorithm based on LDE WiFi indoor positioning method according to claim 4, characterized in that a geodesic minimum spanning tree algorithm for indoor signal coverage map obtained in step two Radio Map intrinsic dimension analysis performed by the formula:
Figure CN103079269AC00021
实现的;式中:dintHnsic;dim为本征维数分析结果;&表示最小生成树的线性拟合表达式的斜率。 Achieved; wherein: dintHnsic; dim several intrinsic dimension analysis; & represents the minimum spanning tree fit linear slope expression.
6.根据权利要求1所述的基于LDE算法的WiFi室内定位方法,其特征在于步骤四中获得特征变换矩阵V'与生成降维后的信号覆盖图Radio Map*之间的关系为: Radio Map* = Y'.Radio Map。 6. WiFi indoor positioning method based on LDE algorithm according to claim 1, wherein the step of obtaining feature four transformation matrix V 'and generate lower signal coverage map dimensionality Radio relationship between the Map *: Radio Map * = Y'.Radio Map.
7.根据权利要求1所述的基于LDE算法的WiFi室内定位方法,其特征在于采用KNN算法对步骤五获得的信号强度变换值RSS*与步骤四生成的降维后的信号覆盖图RadioMap*进行位置匹配的方法是通过公式: 7. The method is based on LDE WiFi indoor positioning algorithm according to claim 1, characterized in that a KNN algorithm coverage map signal strength conversion value RSS * and Step V was four generating dimensionality reduction performed after RadioMap * position matching method is by the formula:
Figure CN103079269AC00031
实现的; 式中:(Χ',I')为欲定位点的坐标,(Xi, Yi)为第i个近邻点的坐标,i为正整数;κ为KNN算法中近邻点的总数。 Achieved; where: (Χ ', I') to coordinate want to locate a point, (Xi, Yi) is the i-th coordinate neighboring point, i is a positive integer; κ is the total number of neighbor points KNN algorithm.
Açıklama  şu dilden çevrildi: Çince

基于LDE算法的WiFi室内定位方法 WiFi-based indoor positioning LDE algorithmic methods

技术领域 Technical Field

[0001] 本发明涉及一种室内定位方法。 [0001] The present invention relates to a method for indoor location.

背景技术 Background

[0002] 随着无线局域网络在世界范围的飞速发展和移动终端设备的广泛普及,近年来出现了许多室内定位相关的技术和应用。 [0002] With the widespread popularity of wireless local area networks in the world and the rapid development of mobile terminal equipment, in recent years there has been a lot of indoor positioning technology and related applications. 由于多径效应、信号衰减及室内定位环境的复杂性,基于传统的信号传播模型的室内定位方法难以达到高精度的室内定位要求。 Due to multipath effects, attenuation and complexity of indoor positioning environment, it is difficult to achieve high accuracy indoor location-based indoor positioning method requires conventional signal propagation model. 基于到达时间(Time of Arrival)、到达时间差(Time Difference of Arrival)或到达角度(Angles ofArrival)等定位方法虽然可以基本满足定位精度需求,然而都需要定位终端有额外的硬件设备支持,具有较大局限性,从而导致基于上述几类定位方法的室内定位系统没有得到普及。 Based on the time of arrival (Time of Arrival), arrival time difference (Time Difference of Arrival) or angle of arrival (Angles ofArrival) method, although other location positioning accuracy can basically meet the demand, however, we will need to locate the terminal has additional hardware support, with greater limitations, resulting in not been universal indoor localization system based on the above-mentioned types of positioning method.

[0003] 目前,基于WLAN位置指纹(Finger Print)的WiFi室内定位方法得到了广泛应用。 [0003] Currently, WLAN based on the location of the fingerprint (Finger Print) of WiFi indoor positioning method has been widely used. 该方法的网络构建方法成本低廉,其使用2.4GHz ISMdndustrial Science Medicine)公共频段且无需在现有设施之上添加定位测量专用硬件。 The method of construction of a network of low cost, using 2.4GHz ISMdndustrial Science Medicine) public band and positioning measurement without adding dedicated hardware on existing facilities. 只需要通过移动终端的无线网卡及相应软件测量接收到的接入点(Access Point, AP)的信号强度(Received SignalStrength, RSS),由此来构建网络信号覆盖图(Radio Map),进而通过匹配算法来预测移动用户所处位置的坐标,或相对位置。 Only received through the wireless network card of the mobile terminal and an access point corresponding to the measurement software (Access Point, AP) signal strength (Received SignalStrength, RSS), to thereby construct the network signal coverage map (Radio Map), and further by matching algorithm to predict the coordinates of the mobile user's location, or relative position.

[0004] 然而通过该方式建立的Radio Map包含有庞大的数据信息,且随着定位区域扩大,Radio Map可能(依据定位匹配方式及算法选择)呈指数形势增长。 [0004] However, the manner established by Radio Map contains a large data information, and with the expansion of the location area, Radio Map may (by targeting and matching algorithm selection) exponential growth situation. 获得尽可能多的相关数据特征信息对于整个系统来说会提升定位精度,但是处理大量的特征信息增加算法开销,定位算法无法在处理能力有限的移动终端上有效运行,同时某些特征信息可能是对于定位没有作用甚至有负面作用,致使匹配效率降低,从而导致匹配定位算法的实现变得更加复杂,并且定位精度下降。 Get as much relevant data attribute information for the entire system, will increase the positioning accuracy, but a large number of characteristic information processing algorithms to increase spending, positioning algorithm can not operate effectively on a limited processing capability of the mobile terminal, while some feature information may be For positioning without effect even have a negative effect, resulting in reduced matching efficiency, thus leading to the matching location algorithm becomes more complicated, and positioning accuracy.

[0005] 当AP的数目增加及定位的参考点(Reference Point)增加时,Radio Map的数据信息增加。 [0005] When the number of AP and the positioning of the reference point (Reference Point) increase, increase Radio Map data information. 此时,Radio Map中代表的AP数目的信息表示了数据的维数。 In this case, AP Radio Map information on the number of representatives expressed the dimension of the data. 因此,当AP数目增加,Radio Map就变成了高维数据。 Thus, when increasing the number of AP, Radio Map it becomes a high-dimensional data. 为减轻处理高维数据的负担,降维算法是有效的解决方法之一。 To alleviate the burden of high-dimensional data dimensionality reduction algorithm is an effective solution. 高维数据可能包含很多特征,这些特征都在描述同一个事物,这些特征一定程度上是紧密相连的。 High-dimensional data may contain many characteristics that are describing the same thing, to some extent, these characteristics are closely linked. 如当从各个角度对同一个物体同时拍照时,得到的数据就含有重叠的信息。 As when viewed from different angles on the same object at the same time taking pictures, the data obtained to contain overlapping information. 如果能得到这些数据的一些简化的不重叠的表达,将会极大地提高数据处理运行的效率并一定程度上提高准确度。 If you can get some simplified expression does not overlap the data, will greatly improve the efficiency of data processing operation and improve the accuracy of a certain degree. 降维算法的目的也正是在于提高高维数据的处理效率。 Dimensionality reduction algorithm is exactly the purpose is to improve the processing efficiency of high-dimensional data.

[0006] 除了可以简化数据使其能够高效处理外,降维方法还可以实现数据可视化。 [0006] In addition to simplifying the data so that it can efficiently handle external dimension reduction method may also be data visualization. 由于很多统计学的和机器学习算法对于最优解的准确性很差,降维的可视化应用可以令用户能够实际看到高维数据的空间结构和算法输出的能力,具有很强的应用价值。 Since many statistical and machine learning algorithms for optimal accuracy is poor, lower-dimensional visualization applications can make the user can actually see the output capacity of the spatial structure and algorithm of high dimensional data, it has a strong value.

[0007]目前有很多基于不同目的的降维算法,包括有线性与非线性降维算法。 [0007] There are a lot of dimension reduction algorithm based on different purposes, including linear and nonlinear dimensionality reduction algorithm. 其中PCA(Principal Component Analysis)和LDA(Linear Discriminant Analysis)是典型的线性降维算法。 Wherein the PCA (Principal Component Analysis) and LDA (Linear Discriminant Analysis) is a typical linear dimension reduction algorithm. 这一类算法对于具有线性结构的高维数据有着良好的处理结果,但对于非线性结构的高维数据没有好的结果。 This class of algorithm for high-dimensional data with a linear structure with good results, but for non-linear high-dimensional data structure is not a good result. 典型的非线性降维算法以流形学习(ManifoldLearning)算法。 A typical nonlinear dimensionality reduction algorithm manifold learning (ManifoldLearning) algorithm. 2000年Science杂志上同一期发表了3篇有关于流形学习算法中提出了 2000 on the same issue of Science magazine published three on manifold learning algorithms have been proposed

2 种经典的流形学习算法:LLE (Local Linear Embedding)及ISOMAP (Isometric Mapping)。 Two kinds of classic manifold learning algorithms: LLE (Local Linear Embedding) and ISOMAP (Isometric Mapping). 由此,各种基于不同的准则的流形学习算法被提出并有一部分流形学习算法应用于图像处理方面。 Thus, a variety of manifold learning algorithm based on different criteria have been proposed and partially manifold learning algorithms for image processing.

[0008]目前的WiFi室内定位方法存在的Radio Map数据库大、在线定位阶段计算复杂度高难以在移动终端实现、定位的实时性差等问题。 [0008] The current WiFi indoor positioning methods exist Radio Map database is large, on-line positioning stage is difficult to achieve high computational complexity in the mobile terminal, real-time and poor targeting.

发明内容 DISCLOSURE

[0009] 本发明是为了解决现有的WiFi室内定位方法的定位的实时性差的问题,从而提供一种基于LDE算法的WiFi室内定位方法。 [0009] The present invention is made to solve the problem of poor real-time positioning of the existing WiFi indoor positioning methods, thereby providing an algorithm based on LDE WiFi indoor positioning methods.

[0010] 基于LDE算法的WiFi室内定位方法,它由以下步骤实现: [0010] LDE algorithm based WiFi indoor positioning method, which consists of the following steps to achieve:

[0011] 步骤一、针对室内环境布置N个接入点AP,确保所述环境中任意一点被一个或一个以上的接入点AP发出的信号覆盖,所述N个接入点AP组成WiFi网络;在所述室内环境中均匀设置Nkp个参考点;N和Nkp均为正整数; [0011] Step one for the indoor environment layout of the N AP, and to ensure coverage at any point in the environment is one or more than one access point AP issued the N composition WiFi network access points AP ; uniform set Nkp reference point in the indoor environment; N and Nkp are positive integers;

[0012] 步骤二、选取一个参考点为坐标原点建立二维直角坐标系,获得Nkp个参考点在该二维直角坐标系中的坐标位置,在离线阶段中在每个参考点上利用信号接收机采集来自每一个接入点AP的信号强度RSS值,并作为该接入点AP的位置特征信息;并根据N个接入点AP的位置特征信息构建室内信号覆盖图Radio Map ; [0012] Step two, pick a reference point for the establishment of a two-dimensional coordinate origin of the Cartesian coordinate system, the coordinate position obtained Nkp reference point in the two-dimensional Cartesian coordinate system, in the offline stage, using the signal received at each reference point machine collecting signal strength RSS values from each access point AP, and as the location of the access point AP feature information; and build indoor coverage map Radio Map N access points based on the position of AP feature information;

[0013] 步骤三、采用本征维数估计算法对步骤二获得的室内信号覆盖图Radio Map进行本征维数分析,获得本征维数分析结果; [0013] Step three, using the intrinsic dimension estimation algorithm for step two indoor signal coverage maps were obtained by Radio Map intrinsic dimension analysis to obtain intrinsic dimension analysis;

[0014] 步骤四、根据步骤三获得的本征维数分析结果采用LED算法将室内信号覆盖图Radio Map内的所有参考点降维至本征维数,获得特征变换矩阵,并生成降维后的信号覆盖图Radio Map*; [0014] Step four, according to the intrinsic dimension analysis results obtained in step three LED algorithm using the reference point for all indoor signal coverage map Radio Map dimensionality reduction within the intrinsic dimension to obtain characteristic transformation matrix, and after generating dimensionality reduction The coverage map Radio Map *;

[0015] 步骤五、在在线阶段,测量室内环境中欲定位点的信号强度RSS值,并将该信号强度RSS值与步骤四获得的特征变换矩阵相乘,获得信号强度变换值RSS* ; [0015] Step five, in-line stage, measuring the indoor environment For setpoint signal strength RSS value and the signal strength RSS feature value obtained in step four transformation matrix is multiplied to obtain the signal strength conversion value RSS *;

[0016] 步骤六、采用KNN算法对步骤五获得的信号强度变换值RSS*与步骤四生成的降维后的信号覆盖图Radio Map*进行位置匹配,获得欲定位点的位置坐标,完成欲定位点的室内定位。 [0016] Step 6 KNN algorithm uses signal strength coverage map conversion value RSS * and Step V was generated by four low dimensional position of Radio Map * match, I want to get the position coordinates of the anchor point to complete the desire positioning indoor positioning points.

[0017] 步骤三中采用本征维数估计算法为特征值估计法。 [0017] Step three adopted intrinsic dimension estimation of eigenvalues estimation.

[0018] 步骤三中采用本征维数估计算法为包数估计法。 [0018] Step three adopted intrinsic dimension estimation method to estimate the number of packets.

[0019] 步骤三中采用本征维数估计算法为测地线最小生成树算法。 [0019] Step three adopted intrinsic dimension estimation algorithm of geodesic minimum spanning tree algorithm.

[0020] 采用测地线最小生成树算法对步骤二获得的室内信号覆盖图Radio Map进行本征维数分析是通过公式: [0020] The geodesic minimum spanning tree algorithm for indoor signal coverage map obtained in step two Radio Map intrinsic dimension analysis performed by the formula:

1 1

[0021] intrinsic dim = Z [0021] intrinsic dim = Z

I — Cl I - Cl

[0022] 实现的;式中:dintHnsic;dim为本征维数分析结果;a表示最小生成树的线性拟合表达式的斜率。 [0022] implemented; Where: dintHnsic; dim-oriented analysis of the number of intrinsic dimension; a minimum spanning tree represents the slope of the linear fit of expression.

[0023] 步骤四中获得特征变换矩阵与生成降维后的信号覆盖图Radio Map*之间的关系为: [0023] Step Four characteristics obtained transformation matrix and generate low dimensional signal coverage maps Map * Radio relationship between the:

Figure CN103079269AD00061

[0025] 采用KNN算法对步骤五获得的信号强度变换值RSS*与步骤四生成的降维后的信号覆盖图Radio Map*进行位置匹配的方法是通过公式: [0025] The KNN algorithm signal strength conversion value RSS * and Step V was four drop coverage map generated after Radio Victoria Map * position matching method is by the formula:

[0026] (λΛ/)= — [0026] (λΛ /) = -

[0027] 实现的; [0027] achieved;

[0028] 式中:U',y')为欲定位点的坐标,(Xi,Yi)为第i个近邻点的坐标,i为正整数;K为KNN算法中近邻点的总数。 [0028] where: U ', y') who desire to locate the coordinates of points, (Xi, Yi) is the i-th coordinate neighboring point, i is a positive integer; K is the number of KNN algorithm neighbor points.

[0029] 本发明的WiFi室内定位实时性高。 [0029] WiFi indoor positioning the present invention real-time high. 同时,本发明采用LDE算法将Radio Map降维至本征维数,降低了现有WiFi室内定位方法中存在的Radio Map数据量大,以及降低了在线定位阶段计算复杂度,使其易于在移动终端实现。 Also, the present invention uses LDE Radio Map dimensionality reduction algorithm to intrinsic dimension, reducing the amount of data existing WiFi Radio Map indoor positioning methods exist, and reduce the computational complexity of the online positioning stage, to make it easy on mobile terminal implementation.

附图说明 Brief Description

[0030] 图1是具体实施方式一中所述的实验场景示意图。 [0030] Figure 1 is an experimental scene in the specific embodiment of FIG. 图2是本发明方法的信号流程示意图。 Figure 2 is a signal flow of the method of the present invention. Fig.

具体实施方式 DETAILED DESCRIPTION

[0031] 具体实施方式一、结合图2说明本具体实施方式,基于LDE算法的WiFi室内定位方法,它由以下步骤实现: [0031] A specific embodiment, described in conjunction with Figure 2 of the present embodiment, the LDE algorithm based WiFi indoor positioning method, which is implemented by the following steps:

[0032] 步骤一、针对室内环境布置`N个接入点AP,确保所述环境中任意一点被一个或一个以上的接入点AP发出的信号覆盖,所述N个接入点AP组成WiFi网络;在所述室内环境中均匀设置Nkp个参考点;N和Nkp均为正整数; [0032] Step one for the indoor environment layout `N an AP, and to ensure coverage at any point in the environment is one or more than one access point AP issued, the composition of N WiFi access points AP network; uniform set Nkp reference point in the indoor environment; N and Nkp are positive integers;

[0033] 步骤二、选取一个参考点为坐标原点建立二维直角坐标系,获得Nkp个参考点在该二维直角坐标系中的坐标位置,在离线阶段中在每个参考点上利用信号接收机采集来自每一个接入点AP的信号强度RSS值,并作为该接入点AP的位置特征信息;并根据N个接入点AP的位置特征信息构建室内信号覆盖图Radio Map ; [0033] Step two, pick a reference point for the establishment of a two-dimensional coordinate origin of the Cartesian coordinate system, the coordinate position obtained Nkp reference point in the two-dimensional Cartesian coordinate system, in the offline stage, using the signal received at each reference point machine collecting signal strength RSS values from each access point AP, and as the location of the access point AP feature information; and build indoor coverage map Radio Map N access points based on the position of AP feature information;

[0034] 步骤三、采用本征维数估计算法对步骤二获得的室内信号覆盖图Radio Map进行本征维数分析,获得本征维数分析结果; [0034] Step three, using the intrinsic dimension estimation algorithm for step two indoor signal coverage maps were obtained by Radio Map intrinsic dimension analysis to obtain intrinsic dimension analysis;

[0035] 步骤四、根据步骤三获得的本征维数分析结果采用LED算法将室内信号覆盖图Radio Map内的所有参考点降维至本征维数,获得特征变换矩阵,并生成降维后的信号覆盖图Radio Map*; [0035] Step four, according to the intrinsic dimension analysis results obtained in step three LED algorithm using the reference point for all indoor signal coverage map Radio Map dimensionality reduction within the intrinsic dimension to obtain characteristic transformation matrix, and after generating dimensionality reduction The coverage map Radio Map *;

[0036] 步骤五、在在线阶段,测量室内环境中欲定位点的信号强度RSS值,并将该信号强度RSS值与步骤四获得的特征变换矩阵相乘,获得信号强度变换值RSS* ; [0036] Step five, in-line stage, measuring the indoor environment For setpoint signal strength RSS value and the signal strength RSS feature value obtained in step four transformation matrix is multiplied to obtain the signal strength conversion value RSS *;

[0037] 步骤六、采用KNN算法对步骤五获得的信号强度变换值RSS*与步骤四生成的降维后的信号覆盖图Radio Map*进行位置匹配,获得欲定位点的位置坐标,完成欲定位点的室内定位。 [0037] Step 6 KNN algorithm uses signal strength coverage map conversion value RSS * and Step V was generated by four low dimensional position of Radio Map * match, I want to get the position coordinates of the anchor point to complete the desire positioning indoor positioning points.

[0038] 步骤三中采用本征维数估计算法为特征值估计法、包数估计法或测地线最小生成树算法。 [0038] Step three adopted intrinsic dimension estimation of eigenvalues estimation, the number of packets estimation or geodesic minimum spanning tree algorithm. [0039] 采用测地线最小生成树算法对步骤二获得的室内信号覆盖图Radio Map进行本征维数分析是通过公式: [0039] The geodesic minimum spanning tree algorithm for indoor signal coverage map obtained in step two Radio Map intrinsic dimension analysis performed by the formula:

[0040] [0040]

Figure CN103079269AD00071

[0041] 实现的;式中:dintHnsic;dim为本征维数分析结果;a表示最小生成树的线性拟合表达式的斜率。 [0041] implemented; Where: dintHnsic; dim-oriented analysis of the number of intrinsic dimension; a minimum spanning tree represents the slope of the linear fit of expression.

[0042] 步骤四中获得特征变换矩阵V'与生成降维后的信号覆盖图Radio Map*之间的关系为: [0042] Step Four characteristics obtained transformation matrix V 'and generate lower signal coverage map dimensionality Map * Radio relationship between the:

[0043] Radio Map* = Y'.Radio Map。 [0043] Radio Map * = Y'.Radio Map.

[0044] 采用KNN算法对步骤五获得的信号强度变换值RSS*与步骤四生成的降维后的信号覆盖图Radio Map*进行位置匹配的方法是通过公式: [0044] The KNN algorithm signal strength conversion value RSS * and Step V was four drop coverage map generated after Radio Victoria Map * position matching method is by the formula:

[0045] [0045]

Figure CN103079269AD00072

[0046] 实现的; [0046] achieved;

[0047] 式中:U',y')为欲定位点的坐标,(Xi,Yi)为第i个近邻点的坐标,i为正整数;K为KNN算法中近邻点的总数。 [0047] where: U ', y') who desire to locate the coordinates of points, (Xi, Yi) is the i-th coordinate neighboring point, i is a positive integer; K is the number of KNN algorithm neighbor points.

[0048] 本实施方式中,Radio Map的本征维数的获取通过下述步骤实现: [0048] In the present embodiment, the intrinsic dimension Radio Map is available through the following steps to achieve:

[0049] 本征维数是对于高维数据进行本征空间维数及空间重建所需最小的独立变量的个数。 [0049] intrinsic dimension of high-dimensional data for the intrinsic spatial dimension and spatial reconstruction of the required minimum number of independent variables. 在具体实际计算中,由于高维数据的本征并不明显,通常不是寻求得到确切的本征维数,而是寻求估计本征维数的可信取值。 In particular the actual calculation, due to the high dimensional data intrinsic is not obvious, it is not usually seek to get the exact intrinsic dimension, but seek credible estimate of the intrinsic value of dimension. 具体的说,给定一个来自高维空间的样本,本征维数估计算法的中心任务和重要内容就是通过这些样本数据来确定这个高维结构的本征维数。 Specifically, given a sample from a high-dimensional space, the intrinsic dimension estimation central task and an important part of the algorithm is through these sample data to determine the intrinsic dimension dimensional structure.

[0050] Radio Map的本征维数的估计是LDE算法的重要输入参数,这关系到降维的结果是否能够代表Radio Map的高维空间的特征,因此准确有效的本征维数的估计至关重要。 [0050] Radio Map estimate of intrinsic dimension is an important input parameter LDE algorithm, whether it relates to the result of lower-dimensional feature Radio Map can represent a high-dimensional space, so an accurate estimate of the effective intrinsic dimension to crucial. 目前,常用本征维数估计算法分为两类:局部估计与全局估计。 At present, the common intrinsic dimension estimation algorithm is divided into two categories: local and global estimate estimate. 在本专利中,采用全局部算法估计对Radio Map的本征维数进行估计,并作为LDE算法的输入变量。 In this patent, all-local algorithm estimates of intrinsic dimension Radio Map to estimate, and as input variables LDE algorithm. 本专利中采用测地距最小生成树算法(Geodesic Minimum Spanning Tree,GMST)对Radio Map的本征维数进行估计。 This patent uses geodesic distance minimum spanning tree algorithm (Geodesic Minimum Spanning Tree, GMST) for intrinsic dimension Radio Map estimated. 下面对GMST算法的理论进行分析。 Next, the theory GMST algorithm analysis.

[0051] 测地线最小生成树(GMST)估计是基于测地线最小生成树的长度函数强烈依赖于本征维数d的。 [0051] Geodesic minimum spanning tree (GMST) estimates are based on the minimum spanning tree geodesic length function is strongly dependent on the intrinsic dimension d of. GMST是指定义在数据集X上的近邻曲线的最小生成树。 GMST is a defined neighborhood graph in the dataset X on the minimum spanning tree. GMST的长度函数是在测地线最小生成树中所有边缘对应的欧氏距离之和。 GMST length function is geodesic minimum spanning tree all edges corresponding to the sum of the Euclidean distance.

[0052] 与ISOMAP相似,GMST估计在数据集X上构造一条近邻曲线G,其中,在X内每一个数据点Xi都和它的k个近邻力,相连接。 [0052] and ISOMAP similar, GMST estimated to construct a graph G neighbors in the dataset X, wherein X within each data point Xi are k nearest neighbors and its power, and connected. 测地线最小生成树T定义为X上的最小曲线,它具 Geodesic minimum spanning tree T is defined as the minimum curve X, it has

有长度: There Length:

[0053] [0053]

Figure CN103079269AD00073

[0054] 其中,T是曲线G的所有子树集合,e是树T的一个边缘,ge是边缘e对应的欧氏距离。 [0054] where, T is the graph G all subtrees set, e is an edge of the tree T, ge edge e corresponds to the Euclidean distance. 在GMST估计中,一些子集JC: I由各种大小m组成,并且子集A的GMST的长度L (A)也需要计算。 In GMST estimate, some subset JC: I m composed by a variety of sizes, and a subset of GMST A length L (A) can also be calculated. [0055] 理论上, [0055] In theory,

Figure CN103079269AD00081

是线性的,从而可以由:y = ax+b这种形式的函数来估计,通过最小二乘法可以估算出变量a和b。 Is linear, so that may be made: estimated y = ax + b This form of the function, can be estimated by the least squares method the variables a and b.

[0056] 可以证明:由a的估算值和d=1/1-a能够得到本征维数的估计。 [0056] can be proved by: a estimates and d = 1/1-a can be estimated intrinsic dimension.

[0057] 由GMST算法给出本征维数d的表达式为式(2)所示。 [0057] From GMST algorithm gives intrinsic dimension d of expression for the formula (2). 本征维数d是LDE算法的另一个重要的输入参数。 Intrinsic dimension d is another important LDE algorithm input parameters.

[0058] d =1/1-a [0058] d = 1/1-a

[0059] 运用LDE算法实现对Radio Map进行降维并获取特征权值矩阵过程通过下述步骤实现: [0059] using LDE algorithm to reduce the dimension of the Radio Map feature and gain weight matrix process realized by the following steps:

[0060] LDE算法是基于类间散度及类内散度最大化的一种流形学习算法。 [0060] LDE algorithm is to maximize a manifold learning algorithms based on internal inter-class scatter and class scatter. 在对LDE算法 In the algorithm of LDE

进行理论分析之前对LDE算法给定的输入数据做如下说明:输入高维数据点 Prior to the theoretical analysis of LDE algorithm given input data described as follows: Input high-dimensional data point

Figure CN103079269AD00082

数据点Xi的类标记为Yi ∈{1,2,…,P},其中P表示将高维数据划分为P个子流形,即将输入的高维数据分成P类。 Xi class data point is marked as Yi ∈ {1,2, ..., P}, where P represents the high-dimensional data into P sub-manifold, high-dimensional data is about to enter into P classes.

[0061] 将输入的高维数据表示成矩阵的形式:X = [x1; X2,…,Xffl]∈Rnx'从矩阵表示的形式来看,矩阵中的列代表一个高维数据点。 [0061] high-dimensional representation of the input data in the form of a matrix: X = [x1; X2, ..., Xffl] ∈Rnx 'form from the matrix point of view, matrix column represents a high-dimensional data points. 下面结合错误! The following combination of error! 未找到引用源。 Reference source not found. 所示的LDE算法流程来对其算法理论进行推导。 LDE algorithm flow shown to derive its theory of algorithms.

[0062] 构造邻接图:根据高维数据点的类标记信息及其近邻关系构造无方向图G及G'。 [0062] constructed adjacent graph: high-dimensional data based on the class mark points of information and its neighbor relationship structure undirected graph G and G '. 其中近邻关系是采用KNN算法给出的准则,即选择数据点最近的K个点作为其邻居,G表示当Xi与Xj的类标记信息Ji≠yj时且Xp Xj互为K近邻关系;G'示当Xi与Xj的类标记信息关y」时且X1、χ」互为K近邻关系。 Wherein the neighbor relationship is the use of the guidelines given by KNN algorithm, choose the most recent data point K points as its neighbor, G represents and Xp Xj K neighboring relationship to each other when Xi and Xj class stamp information Ji ≠ yj; G ' shows when Xi and Xj class mark information about y "when and X1, χ" mutual relationship K neighbors.

[0063] 计算权值矩阵:根据(I)构造的邻接图采用类高斯函数进行权值矩阵的计算。 [0063] calculate the weight matrix: According to the (I) structure adjacent diagram is calculated using the Gaussian function weighted matrix. 其表达式为(3)所示。 Its expression is (3). 公式(3)中&表示近邻点Xi与\之间的权值, Equation (3) & is the weight and \ between the neighbor points Xi,

Figure CN103079269AD00083

为近邻点Xi与\之间的范数距离,采用矩阵方式计算范数距离,t为权值归一化参数。 Is the norm distance \ between the neighboring point Xi, calculated using the matrix norm distance, t is the weight normalized parameters.

Figure CN103079269AD00084

[0065] 计算嵌入结果:根据LDE算法的目标——最大化类间散度地同时最小化类内散度。 [0065] calculation nested result: LDE algorithm based on target - between class scatter to maximize while minimizing divergence within the class. 散度采用表示同类数据点及不同类的范数距离表示。 Divergence from the norm of using similar data indicate points and different types of representation. 由LDE算法的目标可以得出其相应的优化目标函数,如式所示。 By the target LDE algorithm we can draw the corresponding objective function, such as formula. [0066] [0066]

Figure CN103079269AD00091

[0067] 根据式(4)给出的优化目标函数作以下分析: [0067] According to equation (4) the objective function is given for the following analysis:

[0068] 根据矩阵范数的计算式:IMII = YAi,计算式表示为矩阵A的矩阵范数的计算方 [0068] According to the norm of the matrix formula: IMII = YAi, the calculation formula for the norm of calculating the number of matrix A

法,计算式给出的方法与矩阵的迹的计算式一致,即: Method, given the method of calculating the trace of the matrix formula is consistent, namely:

Figure CN103079269AD00092

由此式(4)可以表示为矩阵的迹的计算方式: Thus formula (4) can be expressed as the trace of a matrix is calculated:

[0069] [0069]

Figure CN103079269AD00093

[0070] 式(5)可以简化为: [0070] Formula (5) can be simplified as:

[0071] [0071]

Figure CN103079269AD00094

[0072]由矩阵迹的计算的标量性质及权值元素均为实数,可以将式(6)简化为: [0072] calculated by the matrix trace of scalar nature and weight elements are real numbers, can be of the formula (6) reduces to:

[0073] [0073]

Figure CN103079269AD00095

[0074] 根据简单的数学关系,可以将式(7)简化为: [0074] According to a simple mathematical relationship can be of formula (7) reduces to:

[0075] J(V) = 2tr{VT[X(D/ -Ψ )XT]V} (8) [0075] J (V) = 2tr {VT [X (D / -Ψ) XT] V} (8)

[0076] 式⑶中:X为输入数据,λ,V为特征值与特征向量,W,Ψ分别为G及Gi对应的权值矩阵,D及D'均为对角阵,其对角元素可以由式(9)表示。 [0076] ⑶ the formula: X is the input data, λ, V for the eigenvalues and eigenvectors, W, Ψ G and Gi respectively corresponding weight matrix, D and D 'are diagonal, their diagonal elements may be represented by the formula (9).

[0077] [0077]

Figure CN103079269AD00096

[0078] 根据式⑶的推导方式,同理可以将⑷中的约束条件写成如式⑶相似的形式,由此,可以将(4)表示为如下形式: [0078] Derivation of the equation ⑶, ⑷ empathy can be written in the constraint formula ⑶ similar form, makes it possible to (4) is expressed as follows:

[0079] [0079]

Figure CN103079269AD00097

[0080] 对式(10)应用拉格朗日(Lagrange)乘数法,可以得出式(11)所示: [0080] The formula (10) Lagrange (Lagrange) Multiplier Method, can be derived formula (11) as follows:

[0081] X(D' -r )Χτν = λ X(DW)XTv (11) [0081] X (D '-r) Χτν = λ X (DW) XTv (11)

[0082] 对式(11)进行广义特征值分解,得出其特征值分解的特征值及特征向量,表示为:λ =IIApX2,..., λ η]τ,其对应的特征向量为:V = [V1, ν2,…,νη]τ。 [0082] The formula (11) generalized eigenvalue decomposition Eigenvalue Decomposition obtain its eigenvalues and eigenvectors, expressed as: λ = IIApX2, ..., λ η] τ, the feature vector corresponding to: V = [V1, ν2, ..., νη] τ. 取前d个最大的特征值对应的特征向量构成变换矩阵V= [Vl,v2,-,vd]0由LDE算法的输出数据变换方法可以得出,降维后数据为: D take the first largest eigenvalue corresponding eigenvectors transform matrix V = [Vl, v2, -, vd] 0 can be derived from the output of the data conversion method LDE algorithm, data dimension reduction after:

[0083] Zi = V1Xi (12)[0084] 式(12)中,Zi表示输入高维数据点Xi变换后的低维输出数据。 [0083] Zi = V1Xi (12) [0084] formula (12), Zi represents the input low-dimensional output data high-dimensional data point Xi transformed.

[0085] 上述的分析是根据LDE算法流程给出的理论分析及说明。 [0085] the above analysis is based on the theory of LDE algorithm flow analysis and explanation given.

[0086] 由式(5)〜(11)给出LDE算法的理论推导。 [0086] theory by the formula (5) to (11) gives the LDE method of derivation. 通过LDE算法可以得出降维后的信号覆盖图及特征变换矩阵,分别记为Radio Map*和V'。 By LDE algorithm can draw coverage map and feature reduction after the transformation matrix, denoted as Radio Map * and V '.

[0087] 在线定位阶段对RSS及KNN匹配定位通过下述步骤实现: [0087] Live positioning stage for RSS and KNN matching location by the following steps to achieve:

[0088] 结合图2的在线阶段所示的流程图对具体实施方式四进行详细说明。 [0088] FIG Live Stage 2 flowcharts shown in four specific embodiments described in detail. 在线阶段,测试点处接收的RSS= [AP1, AP2,…,APn],与特征变换矩阵V'相乘,从而得出降维后的RSS' = [AP1 ,AP2,…APd],其中d表示本征维数。 Live stage, received at the test points RSS = [AP1, AP2, ..., APn], and feature transform matrix V 'multiplied to arrive after dimensionality reduction of RSS' = [AP1, AP2, ... APd], where d It represents the intrinsic dimensionality. 再采用KNN算法实现RSS^与Radio Map*的匹配。 Then using KNN algorithm RSS ^ matches Radio Map * of. 采用与RSS'最近的K个参考点的坐标的平均值作为测试点(X' , j'),其表达式为: Using the RSS 'mean nearest K coordinate reference point as a test point (X', j '), which was expressed as:

[0089] [0089]

Figure CN103079269AD00101

[0090] AP在具体的室内布置及实验过程如下例所示:图1示为哈尔滨工业大学科学园2A栋12层的平面图示意,基于WiFi的室内定位系统就是基于该实验环境下建立。 [0090] AP in specific interior layout and experiment in the following example: Figure 1 shows the Harbin Institute of Technology Science Park, Building 12 layer 2A schematic plan view, WiFi-based indoor positioning system is based on the establishment of this experimental environment. 在实验环境中,总共布置27个AP,AP离房间地面高度为2米。 In a lab environment, a total of arrangement 27 AP, AP room on the ground from a height of two meters. 在离线阶段,在联想V450笔记本上安装NetStumbler软件,在所有参考点的四个不同的方位上连续采样记录AP的100个RSS值,以及AP的相关信息。 In the offline stage, NetStumbler software installed on Lenovo V450 notebook in four different directions all reference points continuous sampling record 100 RSS value AP, and AP information. 将所有的采样点的物理坐标及相应的物理坐标及RSS值存储为定位过程所调用的数据,建立Radio Map。 The data on the physical coordinates of all sampling points and the corresponding physical coordinates and RSS values are stored as locating the called procedure established Radio Map. 在实验环境共有900个参考点,其采样密度为0.5米X0.5米。 A total of 900 experimental environment reference point, the sampling density is 0.5 m X0.5 m. Radio Map作为LDE算法的输入参数及本征维数估计算法的输入数据。 Radio Map LDE algorithm as an input parameter and intrinsic dimension estimation algorithm input data.

Patent Atıfları
Alıntı Yapılan Patent Dosya kabul tarihi Yayın tarihi Başvuru sahibi Başlık
CN102348160A *15 Tem 20118 Şub 2012中国电信股份有限公司Multimode signal-based positioning method, system and positioning platform
CN102419180A *2 Eyl 201118 Nis 2012无锡智感星际科技有限公司Indoor positioning method based on inertial navigation system and WIFI (wireless fidelity)
US20110164522 *14 Mar 20117 Tem 2011Skyhook Wireless, Inc.Estimation of Position Using WLAN Access Point Radio Propagation Characteristics in a WLAN Positioning System
Patent Harici Atıflar
Referans
1 *邓志安: "《基于学习算法的WLAN室内定位技术研究》", 《哈尔滨工业大学》, 1 June 2012 (2012-06-01)
Referans veren:
Alıntı Yapan Patent Dosya kabul tarihi Yayın tarihi Başvuru sahibi Başlık
CN103648106A *31 Ara 201319 Mar 2014哈尔滨工业大学WiFi indoor positioning method of semi-supervised manifold learning based on category matching
CN103648106B *31 Ara 201322 Mar 2017哈尔滨工业大学一种基于类别匹配的半监督流形学习的WiFi室内定位方法
CN104424276A *30 Ağu 201318 Mar 2015中国电信集团公司Method and device for self-updating fingerprint database based on manifold learning
Sınıflandırma
Uluslararası SınıflandırmaH04W64/00
Yasal Etkinlikler
TarihKodEtkinlikAçıklama
1 May 2013C06Publication
5 Haz 2013C10Entry into substantive examination
13 Nis 2016C02Deemed withdrawal of patent application after publication (patent law 2001)