Zoeken Afbeeldingen Maps Play YouTube Nieuws Gmail Drive Meer »
Inloggen
Gebruikers van een schermlezer: klik op deze link voor de toegankelijkheidsmodus. De toegankelijkheidsmodus beschikt over dezelfde essentiŽle functies, maar werkt beter met je lezer.

Patenten

  1. Geavanceerd zoeken naar patenten
PublicatienummerCN102610000 A
PublicatietypeAanvraag
AanvraagnummerCN 201210065998
Publicatiedatum25 juli 2012
Aanvraagdatum14 maart 2012
Prioriteitsdatum14 maart 2012
Publicatienummer201210065998.4, CN 102610000 A, CN 102610000A, CN 201210065998, CN-A-102610000, CN102610000 A, CN102610000A, CN201210065998, CN201210065998.4
Uitvinders刘 东
Aanvrager江苏钱旺网络科技有限公司
Citatie exporterenBiBTeX, EndNote, RefMan
Externe links:  SIPO, Espacenet
Employee attendance location method based on Wi-Fi (wireless fidelity) technology
CN 102610000 A
Samenvatting
The invention relates to an employee attendance location method based on the Wi-Fi (wireless fidelity) technology, aiming to solve the problems that the signal-strength-based triangulation location process is low in precision due to high affection of external factors to Wi-Fi signals, and application of the Wi-Fi technology to employee management for enterprises is limited due to the difficulty in acquiring the location information of the access point (AP). The employee attendance location method includes: (1), acquiring reference point location signal characteristics including the basic service set identifier (BSSID) of the access point, and building a location fingerprint database; (2) sequencing the RSS (really simple syndication) vectors acquired from to-be-tested points according to strength of signals, searching the previous 8-15 pieces of record with the strongest signals, matching the corresponding fingerprint records from the fingerprint database, and computing the matching fingerprints and the Euclidean distances of the to-be-tested points and estimating the location of the to-be-tested points. The employee attendance location method based on the Wi-Fi technology is used for managing indoor staff and outdoor staff of enterprises in any time.
Claims(3)  vertaald uit het Chinees
1. 一种基于Wi-Fi技术的员工考勤定位的方法,其特征是:该方法包括如下步骤:(1)采集参考点位置的信号特征,包括AP的唯一标识(BSSID)和信号强度,建立一个位置指纹数据库;(2)将待测点所采集到的RSS矢量按照信号强度排序后,找出信号最强的前8-15条记录,然后在指纹数据库中匹配出相应的指纹记录,计算出匹配的指纹和待测点的欧氏距离并估算出待测点的位置。 1. A method for Wi-Fi technology staff attendance based positioning, wherein: the method comprising the steps of: (1) the acquisition of the reference position signal characteristics, including the unique identification of AP (BSSID) and signal strength, build A location fingerprinting database; (2) to be tested point the collected signal strength RSS vectors sorted according to find the strongest signal before 8-15 record, then the corresponding fingerprint matching fingerprint records in the database, calculation the Euclidean distance matching fingerprint and the target point and to estimate the location of the target point.
2.根据权利要求I所述的基于Wi-Fi技术的员工考勤定位的方法,其特征是:步骤(I) 中所述的采集参考点位置的信号特征时扫描Wi-Fi信号30次取平均值。 The method is based on Wi-Fi technology I said staff attendance positioning claim, characterized in that: step (I) collection of reference position signal when the scanning features Wi-Fi signal averaged 30 times value.
3.根据权利要求I或2所述的基于Wi-Fi技术的员工考勤定位的方法,其特征是:步骤(2)中所述的计算出匹配的指纹和待测点的欧氏距离时采用近邻算法:假设待测点接收到的RSS观测值为: The method I or Wi-Fi technology based on staff attendance positioning 2 claim, characterized in that: in the step (2) is calculated according to the matching of fingerprints and the target point of the Euclidean distance adoption neighbor algorithm: Suppose the target point value received RSS observations:
Figure CN102610000AC00021
数据库中的已有记录为 Database has been recorded as
Figure CN102610000AC00022
其中η代表待检测点上检测到的不同AP数;ie [l,Nt],Nt为数据库中的记录数,Ni 代表第i条记录中存储的不同AP数,则近邻算法可表示为: Wherein the different number detected on AP η representatives measuring point; ie [l, Nt], Nt is the number of records in the database, Ni represents the first i records stored in different AP count, the nearest neighbor algorithm can be expressed as:
Figure CN102610000AC00023
其中11S- Si 11代表s和Si之间的欧氏距离。 Which 11S- Si 11 on behalf of the Euclidean distance between s and Si.
Beschrijving  vertaald uit het Chinees

基于Wi-Fi技术的员工考勤定位的方法 The method of Wi-Fi technology staff attendance positioning based

[0001] 技术领域: [0001] Technical Field:

本发明涉及无线网络应用领域,具体涉及一种基于Wi-Fi技术的员工考勤定位的方法。 The present invention relates to the field of wireless networking applications, specifically to a method based on Wi-Fi technology, staff attendance positioning.

[0002] 背景技术: [0002] BACKGROUND:

Wi-Fi (Wireless Fidelity)技术是无线局域网(WLAN)技术-IEEE 80211系列标 Wi-Fi (Wireless Fidelity) technology is a wireless local area network (WLAN) technology -IEEE 80211 series standard

准的商用名称。 Quasi-commercial names. IEEE 802. 11系列标准主要包括IEEE 802. lla/b/g/n 4种。 IEEE 802. 11 series of standards including IEEE 802. lla / b / g / n 4 species. 在开放性区域,Wi-Fi的通信距离可达305 m;在封闭性区域,通信距离为76—122 m。 In an open area, Wi-Fi communication distance up to 305 m; in a closed area, communication distance of 76-122 m. 目前还没有将Wi-Fi技术应用在企业的员工管理的先例,目前的企业普遍实行的考勤方式有指纹考勤、 打卡考勤等,这些考勤方式都无一例外地需要员工现场签到,这就出现了在上下班的高峰时间排队签到的情况,员工较多的企业由于排队等待的时间过长,致使一部分原本没有迟到的员工签到的时间超过了规定的上班时间。 There is no Wi-Fi technology will be used in the management of the company's employees precedent, the current widespread use of business appraisal method fingerprint attendance, time and attendance, etc. These methods are invariably require the attendance of employees on-site attendance, which appeared on peak commuting time queuing to sign, employee more enterprises because of waiting too long, causing part of the staff had no late check-in time exceeds a predetermined working hours. 而对于在外出差的员工就无法考勤,企业也不能随时掌握其动向,形成了企业管理上的盲点。 For employees can not travel abroad and attendance, business can not keep track of their movements, formed a blind spot business management. 也有依赖于全球定位系统(Global Positioning System,GPS )是获取员工远程地理位置信息的应用,但由于卫星信号易受到障碍物的遮挡,GPS和AGPS等卫星定位技术并不适用于室内(商场、办公室等)或高楼林立等有障碍物遮挡的场合。 There is also dependent on the GPS system (Global Positioning System, GPS) is to obtain staff remote geographic information applications, but because satellite signals are vulnerable to obstacles blocking, GPS and AGPS and other satellite positioning technology is not suitable for indoor (shopping malls, office etc.) or high-rise buildings and other obstacles blocking the occasion. 目前比较常用的定位技术有基于到达时差、基于到达角度、还有基于接收信号强度来定位移动终端,由于Wi-Fi信号受外界因素影响较大,故基于信号强度的三角定位法的精度较差,而且获取接入点(Access Point, AP)的位置信息也比较困难,限制了Wi-Fi技术在企业的员工管理方面的应用。 Now commonly used TDOA based positioning technology, based on the angle of arrival, and based on the received signal strength to locate the mobile terminal, because the Wi-Fi signal is influenced by external factors, it is less accurate triangulation method based on signal strength and obtain access point (Access Point, AP) location information is also more difficult, limiting the Wi-Fi technology in the management of the enterprise's employees.

[0003] 发明内容: [0003] SUMMARY:

本发明的目的是针对上述存在的问题提供一种基于Wi-Fi技术的员工考勤定位的方法,定位准确,迅速。 The present invention is directed to the above problems to provide a method for Wi-Fi technology staff attendance based positioning, accurate and rapid.

[0004] 上述的目的通过以下的技术方案实现: [0004] The above object is achieved by the following technical scheme:

基于Wi-Fi技术的员工考勤定位的方法,该方法包括如下步骤: The method of Wi-Fi technology staff attendance positioning based, the method comprising the steps of:

(1)采集参考点位置的信号特征,包括AP的唯一标识(BSSID)和信号强度,建立一个位置指纹数据库; (1) Acquisition of the reference position signal characteristics, including the unique identification of the AP (BSSID) and the signal strength, to establish a position of the fingerprint database;

(2)将待测点所采集到的RSS矢量按照信号强度排序后,找出信号最强的前8-15条记录,然后在指纹数据库中匹配出相应的指纹记录,计算出匹配的指纹和待测点的欧氏距离并估算出待测点的位置。 (2) to be tested point the collected RSS vectors sorted according to signal strength, find the strongest signal before 8-15 record, and then match the fingerprint database in the corresponding fingerprint records, fingerprints and matching calculated Euclidean distance between the target point and to estimate the location of the target point.

[0005] 所述的基于Wi-Fi技术的员工考勤定位的方法,步骤(I)中所述的采集参考点位置的信号特征时扫描Wi-Fi信号30次取平均值。 [0005] The method of Wi-Fi technology staff attendance positioning based on step (I) scanning Wi-Fi signal acquisition reference position signal characteristics in the 30 times the average.

[0006] 所述的基于Wi-Fi技术的员工考勤定位的方法,步骤(2)中所述的计算出匹配的指纹和待测点的欧氏距离时采用近邻算法:假设待测点接收到的RSS观测值为: [0006] The method of Wi-Fi technology staff attendance positioning based on said step (2) is calculated using the nearest neighbor algorithm matching fingerprints and Euclidean distance of the target point: Suppose the target point received The RSS observation is:

Figure CN102610000AD00031

,,数据库中的已有记录为 ,, The database has been recorded as

Figure CN102610000AD00032

其中n代表待检测点上检测到的不同AP数;i G [l,Nt],Nt为数据库中的记录数,Ni 代表第i条记录中存储的不同AP数,则近邻算法可表示为: Wherein the different AP count detected measuring point represent n; i G [l, Nt], Nt is the number of records in the database, Ni represents the first i records stored in different AP count, the nearest neighbor algorithm can be expressed as:

Figure CN102610000AD00041

其中| | S- Si II代表S和Si之间的欧氏距离。 Where | | S- Si II on behalf of the Euclidean distance between S and Si.

[0007] 有益效果: [0007] beneficial effects:

1.本发明采用Wi-Fi技术可以方便地与现有的有线以太网络整合,组网成本低。 1. The present invention uses Wi-Fi technology can be easily with existing wired Ethernet network integration, low cost networking. Wi-Fi 是由接入点(Access Point, AP)和无线网卡组成的无线网络,结构简单,可以实现快速组网,架设费用和程序的复杂性远远低于传统的有线网络。 Wi-Fi access point is (Access Point, AP) and wireless LAN composed of wireless network, a simple structure, can achieve fast network, set up costs and complexity of the program is far lower than traditional wired networks. 两台以上的电脑还可以组建对等网,不需要AP.只需每台电脑配备无线网卡。 Two or more computers can also set up peer, without AP. Just every computer equipped with a wireless network card. AP作为传统的有线网络与无线局域网之间的桥梁,任何一台装有无线网卡的PC都可以通过AP接入有线网络,其工作原理相当于一个内置无线发射器的集线器或者路由器,无线网卡则是负责接收AP所发射信号的客户端。 AP as a bridge between traditional wired networks and wireless LANs, and any PC equipped with a wireless card can access the wired network by AP, it works the equivalent of a built-in wireless transmitter hub or router, wireless network card AP is responsible for receiving the transmitted signal client. 企业可以在预设地点收集可用来作为考勤指标的Wi-Fi指纹信息,员工就可以在其预设点附近方便的完成打卡操作,尤其是在一些电梯比较繁忙的单位,企业可以提供一些人性化的签到机制方便员工签到。 Wi-Fi companies can collect fingerprint information is available in the default location indicators as attendance, employees can easily complete the punch operating near its preset point, especially in some of the busier elevator units, enterprises can provide some humanity facilitate the employees sign the attendance mechanism. 采用Wi-Fi定位系统,记录外勤人员的运动轨迹,然后用Google Map 以图形化的形式重现其运动轨迹,达到企业对外勤人员的考勤。 Using Wi-Fi Positioning System to record the trajectory of field staff, and then use the Google Map to graphically reproduce the form of its trajectory, to foreign enterprises and service personnel in attendance.

[0008] 由于Wi-Fi信号的不确定性以及受外界因素的影响比较大,本发明经过多次实践结果表明,在连续扫面30次左右,Wi-Fi信号趋于稳定,所以本发明采用扫描Wi-Fi信号30 次取平均值后来建立指纹数据库。 [0008] Due to the uncertainty of Wi-Fi signal is affected by external factors and the relatively large, the present invention after several practice results show that about 30 times in a row sweep surface, Wi-Fi signals to stabilize, the present invention employs Wi-Fi signal scanning 30 times averaged later established fingerprint database.

[0009] 由于最近邻法需要遍历整个数据库,并计算出每条指纹和待测点的欧氏距离,然后找出最小值的指纹记录从而估算待测位置,该方法计算量巨大。 [0009] Since the nearest neighbor method needs to traverse the entire database, and calculates the Euclidean distance between each finger and the target point, and then find the minimum value of fingerprint records to estimate the test location, the method for calculating huge amount. 我们所采取的是,将待测点所采集到的RSS矢量按照信号强度排序后,找出信号最强的前N条记录,然后在指纹数据库中匹配出相应的指纹记录,从而只需要计算出匹配的指纹和待测点的欧氏距离并估算出待测点的位置,极大的减少了计算量,为支持海量用户奠定了技术基础。 We have taken is to be tested point the collected RSS vectors sorted according to signal strength, find the strongest signal of the first N records, then the corresponding fingerprint matching fingerprint records in the database, so only need to calculate the Euclidean distance matching fingerprint and the target point and the estimated position of the target point, greatly reducing the amount of computation to support massive user to lay the technical foundation.

[0010] 由于欧氏距离越小,则表示真实距离越近,该值的可信度也越高,故该值所占的权重应该更大。 [0010] Since the smaller the Euclidean distance is, the more nearly true distance, the higher the credibility of the value, so the value should be larger share of the weight. 改进后的算法比普通近邻算法的定位增加了一定的精度。 The improved algorithm adds a certain positioning accuracy than ordinary neighbor algorithm.

[0011] 具体实施方式: [0011] DETAILED DESCRIPTION:

下面结合具体的实施例对本发明的技术方案做进一步说明: Below with specific examples of technical solutions of the present invention is further illustrated:

基于Wi-Fi技术的员工考勤定位的方法,该方法包括如下步骤: The method of Wi-Fi technology staff attendance positioning based, the method comprising the steps of:

(1)扫描Wi-Fi信号30次取平均值,采集参考点位置的信号特征,包括AP的唯一标识(BSSID)和信号强度,建立一个位置指纹数据库; (1) Scan Wi-Fi signal averaged 30 times, the acquisition of the reference position signal characteristics, including the unique identification of the AP (BSSID) and the signal strength, to establish a position of the fingerprint database;

(2)将待测点所采集到的RSS矢量按照信号强度排序后,找出信号最强的前8-15条记录,然后在指纹数据库中匹配出相应的指纹记录,计算出匹配的指纹和待测点的欧氏距离并估算出待测点的位置。 (2) to be tested point the collected RSS vectors sorted according to signal strength, find the strongest signal before 8-15 record, and then match the fingerprint database in the corresponding fingerprint records, fingerprints and matching calculated Euclidean distance between the target point and to estimate the location of the target point. 所述的计算出匹配的指纹和待测点的欧氏距离时采用近邻算法: The calculated matching fingerprints and the target point of the Euclidean algorithm using the nearest neighbor distance:

假设待测点接收到的RSS观测值为: Hypothesis test point value received RSS observations:

Figure CN102610000AD00042

,,数据库中的已有记录为 ,, The database has been recorded as

Figure CN102610000AD00043

其中η代表待检测点上检测到的不同AP数;ie [l,Nt],Nt为数据库中的记录数,Ni 代表第i条记录中存储的不同AP数,则近邻算法可表示为: Wherein the different number detected on AP η representatives measuring point; ie [l, Nt], Nt is the number of records in the database, Ni represents the first i records stored in different AP count, the nearest neighbor algorithm can be expressed as:

Figure CN102610000AD00051

其中11S- Si 11代表s和Si之间的欧氏距离。 Which 11S- Si 11 on behalf of the Euclidean distance between s and Si.

[0012] 普通的近邻算法定位: [0012] Common nearest neighbor Location:

对于匹配的N个参考点,求其平均值估算出该点的位置。 N for matching reference points, find the average position of the point estimate.

[0013] 例如对三个参考点计算的坐标位置(X,y)分别为(latl, lonl), (lat2,lon2), (lat3, 1οη3),则最终估算出来的点为((Iatl+lat2+lat3)/3,(1οη1+1οη2+1οη3)/3) [0013] For example the coordinate position (X, y) of the three reference points are calculated to be (latl, lonl), (lat2, lon2), (lat3, 1οη3), the final out of the point estimate ((Iatl + lat2 + lat3) / 3, (1οη1 + 1οη2 + 1οη3) / 3)

改进后的算法对匹配的参考点进行了新的权重分配,分别计算出该样本点与参考点的欧氏距离,将欧氏距离的倒数作为权重。 The improved algorithm for matching reference point for a new weight distribution, were calculated Euclidean distance to the sample point and the reference point, the reciprocal of the Euclidean distance as weights.

[0014] 例如匹配三个点,计算的欧氏距离值分别为a,b, C。 [0014] For example by matching three points, Euclidean distance values are calculated to be a, b, C. 这三个点的坐标位置(X,y) 分别为(latl,lonl),(lat2, lon2), (lat3,1οη3),则最终估算出来的点为((latl/a + lat2/b + lat3/c)/(1/a+l/b+l/c),(lonl/a + lon2/b + lon3/c)/(1/a+l/b+l/c))。 The three point coordinate position (X, y) were (latl, lonl), (lat2, lon2), (lat3,1οη3), the final point is estimated ((latl / a + lat2 / b + lat3 / c) / (1 / a + l / b + l / c), (lonl / a + lon2 / b + lon3 / c) / (1 / a + l / b + l / c)).

[0015] 以10个AP信号为例,经典的近邻算法含21个基本计算,10减法运算,10乘法运算,I个开方运算,由于经典的近邻算法需要遍历整个数据库,假设指纹数据库中记录了100个指纹数据,因此计算量为100 * 21。 [0015] In 10 AP signal as an example, the classical nearest neighbor 21 basic computing including 10 subtraction, multiplication 10, I a root operation, due to the classical neighbor algorithm needs to traverse the entire database, assuming that records fingerprint database a fingerprint data 100, and therefore the amount of calculation is 100 * 21. 本发明经过改良后的算法,先根据待测点的RSS 矢量,匹配出指纹矢量后计算量变为10 * 21 + Tn (其中Tn为根据待测点RSS筛选指纹记录所需要的计算量),可以确定的是,Tn的计算量大大小于90个欧氏距离的计算量。 The present invention is improved through the algorithm, first tested in accordance RSS vector points, matching a fingerprint vector calculation amount becomes 10 * 21 + Tn (Tn where computation for the screening of fingerprint records based on the desired target point RSS), you can determined that, Tn calculations on large size calculation 90 Euclidean distance. 改进的近邻算法的均方根误差为9. 68m,最大定位误差为29.3m,在扫描到无线信号后,整个定位过程在I秒以内。 RMSE improved neighbor algorithm is 9. 68m, maximum positioning error of 29.3m, after scanning the wireless signal, I whole positioning within seconds.

Patentcitaties
Geciteerd patent Aanvraagdatum Publicatiedatum Aanvrager Titel
CN101191833A *12 dec 20074 juni 2008北京航空航天大学Radio frequency recognition indoor position finding and detection method based on receiving signal intensity
CN101505546A *26 feb 200912 aug 2009清华大学Mine environment probing and search and rescue emergency response system based on wireless sensor network
CN101551275A *30 april 20097 okt 2009上海航遥信息技术有限公司;中国科学院上海技术物理研究所Technical method of vehicular multispectral scanner for monitoring industrial warm discharge water
CN101639527A *3 sept 20093 feb 2010哈尔滨工业大学K nearest fuzzy clustering WLAN indoor locating method based on REE-P
CN101651951A *15 sept 200917 feb 2010哈尔滨工业大学Establishing method and positioning method of indoor positioning network of support vector machine based on WLAN
CN101818653A *31 dec 20091 sept 2010深圳市翌日科技有限公司Received signal strength based real-time location method for people under mine
CN101883424A *23 april 201010 nov 2010哈尔滨工业大学WLAN (Wireless Local Area Network) indoor KNN (K-Nearest Neighbor) positioning method based on near-neighbor point number optimization
EP1689126A1 *8 feb 20059 aug 2006Alcatel Alsthom Compagnie Generale D'electriciteLocation service for use in a Wireless LAN
WO2007133967A2 *4 mei 200722 nov 2007Skyhook Wireless, Inc.Estimation of position using wlan access radio propagation characteristics in a wlan positioning system
WO2009115991A1 *17 maart 200924 sept 2009Nxp B.V.System and method of providing network status information of a basic service set for selective access point association
Niet-patentcitaties
Referentie
1 *曾勇: "广义近邻模式分类研究", 《中国博士学位论文全文数据库 信息科技辑》, 15 April 2011 (2011-04-15), pages 138 - 79
Verwijzingen naar dit patent
Citerend patent Aanvraagdatum Publicatiedatum Aanvrager Titel
CN103426208A *23 aug 20134 dec 2013上海同城信息科技有限公司Cell phone attendance system based on WiFi signals
CN104157019A *21 juli 201419 nov 2014惠州Tcl移动通信有限公司Indoor-positioning-technology-based check-in method and system
CN105759240A *19 dec 201413 juli 2016中国移动通信集团吉林有限公司Positioning reference information update method and device
CN106127872A *21 juni 201616 nov 2016北京奇虎科技有限公司Work attendance checking method based on mobile terminal, client side and device
WO2016011697A1 *27 aug 201428 jan 2016惠州Tcl移动通信有限公司Sign-in method and system based on indoor positioning technology
Classificaties
Internationale classificatieH04W84/12, G07C1/10
Juridische gebeurtenissen
DatumCodeGebeurtenisBeschrijving
25 juli 2012C06Publication
26 sept 2012C10Entry into substantive examination
17 april 2013C53Correction of patent for invention or patent application
17 april 2013CORChange of bibliographic data
Free format text: CORRECT: APPLICANT; FROM: JIANGSU QIANWANG NETWORK SCIENCE + TECHNOLOGY CO., LTD. TO: JIANGSU QIANWANG INTELLIGENCE SYSTEM CO., LTD.
29 juli 2015C12Rejection of a patent application after its publication