CN201699986U - Indoor Positioning Device Based on WLAN - Google Patents

Abstract

The utility model relates to the field of indoor positioning equipment based on a wireless local area network. It is an indoor positioning device based on wireless LAN, including a server monitoring module that receives the location information sent by each client positioning module in real time and displays the user's location information on the map in real time, and defines subspaces, location reference points, and paths. The location investigation module of various information and test analysis, real-time tracking, wireless network card setting, etc., calculates the current location of the user and sends the location information to the client positioning module of the server monitoring system; the server monitoring module, location investigation module, customer The terminal positioning modules communicate with each other through a wireless local area network, and the wireless local area network is an 802.11 wireless local area network. The utility model has the advantages of low cost and high positioning accuracy. In a typical office building environment, 80% of the positioning accuracy can be achieved within 3 meters, and 90% of the positioning accuracy is within 5 meters; and the single positioning time is short, and the single positioning time of the system is less than 25 seconds.

Description

Indoor Positioning Device Based on WLAN

technical field

The utility model belongs to the field of communication, in particular to the field of indoor positioning equipment based on a wireless local area network.

Background technique

The emergence and development of satellite navigation and positioning technology has enabled people to have a technical method to obtain the spatial location attributes of things in the vast outdoor space around the world, basically solving the problem of positioning in an outdoor space with good air-to-air conditions, and It has been widely used in the fields of military affairs, transportation, resources and environment, agriculture, animal husbandry and fishery, surveying and mapping, and people's daily life. Although the outdoor space is vast, people's activities are still carried out indoors most of the time. Under the current technical conditions, the standard error of GPS is relatively large compared with the small indoor environment, and because the indoor environment signal is affected The positioning accuracy will be greatly affected by shadowing, and the satellite navigation and positioning technology is still difficult to meet the positioning requirements in the indoor environment. Therefore, for the special indoor environment, it is necessary to study a special positioning method for indoor positioning.

Since the late 1990s, many universities and research institutions have started research on indoor positioning technology. The representative one is the Active Badges project hosted by AT&T Cambridge, which was further improved to Active Bats, Cricke, Microsoft's Easy Living project and Georgia Tech's Smart Floor project, etc. Although the above-mentioned projects have achieved certain results, and some can achieve millimeter-level accuracy, these positioning systems all need to add new hardware, the system deployment is complicated, the maintenance cost is high, and the scalability is poor.

Utility model content

The purpose of the present invention is to solve the above problems and provide an indoor positioning device with low cost, high positioning accuracy, simple system deployment and good scalability.

The utility model is achieved in that:

Indoor positioning equipment based on wireless local area network, which includes a server monitoring module that receives the location information sent by each client positioning module in real time and displays the user's location information on the map in real time, and defines subspaces, location reference points, paths, etc. The location investigation module of various information and test analysis, real-time tracking, wireless network card settings, etc., calculates the current location of the user and sends the location information to the client positioning module of the server monitoring system; server monitoring module, location investigation module, client The positioning modules communicate with each other through the wireless local area network.

The wireless local area network mentioned above is an 802.11 wireless local area network.

The client positioning module adopts a signal strength filtering method when reading signal strength data.

The client positioning module adopts a user location filtering method when processing user location information.

When the location investigation module defines various information such as subspaces, location reference points, paths, etc., the path information of the indoor environment is used to rationally optimize the search subspace of the positioning algorithm, that is, the method of path tracking assistance.

The operating system platform that the server monitoring module operates is Windows XP, Windows 2003 Server; the operating system platform that the position investigation module operates is Windows XP; the operating system platform that the client positioning module operates is Windows XP/2000.

At present, wireless local area networks based on 802.11b/g protocols have been widely distributed in campuses, office buildings and homes. Mobile devices such as handheld computers and notebooks also have built-in wireless network cards. Wireless local area network positioning based on received signal strength (RSS) Positioning is based on the law that the strength of the received signal changes with the distance. Compared with the wireless local area network positioning technology based on signal time of arrival (TOA) and signal angle of arrival (AOA), it does not need to add additional hardware devices for accurate positioning. Time synchronization and angle measurement can make full use of the existing wireless network facilities and expand the application range of the positioning system to buildings and indoors, thereby greatly reducing the cost, so it has become a research hotspot in indoor positioning technology.

The utility model is mainly composed of three major component modules, which are respectively a server monitoring module, a WLAN position investigation module and a client positioning module. These three component modules can communicate with each other through 802.11a/b/g wireless network, and the location fingerprint database is stored in the central database. The functions of each module are described as follows:

(1), the server monitoring module, the operating system platform of which is Windows XP, Windows 2003 Server, the main function is to display the current position of all positioning devices on the screen in real time. The server monitoring system receives the location information sent by each client positioning module in real time through the UPD protocol, and updates the user's location information on the map in real time.

(2), WLAN location investigation module, its running operating system platform is Windows XP, the main functions include: new project, project open, import map, define subspace, define location reference point, define path information, define map scale, investigate Calibration, survey testing, signal strength analysis, precision statistical analysis, error vector analysis, real-time tracking and wireless network card settings, etc.

(3), the client positioning module, the operating system platform of its operation is Windows XP/2000, the main function is to calculate the current location of the user in real time on the handheld device (notebook), and send the location information (x, y) through UDP The protocol is sent to the server monitoring module.

The principle of the technical route mainly adopted by the indoor positioning equipment includes:

(1) Perform filtering and noise reduction processing on the original signal strength data obtained by sampling, that is, signal strength filtering

Due to the complexity of the indoor environment, the signal strength (RSSI) obtained by the wireless network card is subject to noise interference from various factors, such as signal interference from mobile phones and microwave ovens, reflection, refraction, and diffraction caused by walls and ceilings, and between APs. frequency band interference, as well as sudden movement of people and the opening and closing of doors and windows. These random interferences will lead to large variance and noise in the signal strength obtained by sampling, which seriously affects the positioning accuracy of the indoor positioning system. In order to solve this problem, it is necessary to use a filtering algorithm to denoise the original data, reduce the variance, and improve the positioning accuracy of the positioning equipment. Since the noise interference in the indoor environment (such as when someone passes by) will generally reduce the RSSI, the indoor positioning equipment uses the maximum value filtering method to filter, that is, take a maximum value from each continuous value, and filter out the interference. This results in a lower value of RSSI, thereby reducing the variance of RSSI and improving the positioning accuracy of indoor positioning equipment.

(2) Use the filtering algorithm to filter the user position coordinates estimated by the positioning device, that is, user position filtering

In dynamic real-time tracking applications, the positioning device needs to calculate the current location information of the user in about 5 seconds to achieve real-time goals. Therefore, the signal strength sample data read by the wireless network card during real-time tracking is relatively small, for example, only 10 signal strength data are read in 5 seconds (when the scanning frequency is defined as 2 times per second), and various positioning algorithms are directly used at this time When calculating the user's location, the obtained user's location changes greatly, so it is necessary to further improve the performance of the indoor positioning device by means of Kalman filter or particle filter. The positioning device adopts Kalman filter algorithm (Kalman Filter) to filter the user position coordinates estimated by the system. The observation variables of the Kalman filter algorithm are four, namely [x, y, v _x , v _y ].

(3) Use the path information of the indoor environment to reasonably optimize the search subspace of the positioning algorithm, that is, path tracking assistance

In the implementation of the application, the user moves according to a certain path, such as walking from one end of the corridor to the other, entering from one room to another, and it is impossible for the user to pass through the wall. Therefore, position information such as various possible paths of the user can be defined in advance, and then the positioning algorithm can combine the path information to improve the positioning accuracy of the system. The specific method is as follows: first define the subspace information of the indoor environment, and then define the possible path information between these subspaces. The system first calculates the subspace where the current user's location is located. When estimating the user's next location, the system only matches and calculates from the current subspace and the subspace connected to the current subspace, so the position coordinates estimated by the system cannot jump If it is too large, it basically moves according to the path defined in advance; it is even more impossible to pass through the wall, because the path information for passing through the wall does not exist. Moreover, after such optimization, the subspaces to be searched and matched by the positioning algorithm are greatly reduced (originally all subspaces to be searched and matched), and thus the time required for calculating the coordinate position is also greatly reduced.

The deployment of this indoor positioning device can generally be divided into the following six steps:

(1) Determine the venue for indoor positioning and draw a map file of the venue;

(2) Determine the model and quantity of wireless access points (APs), and rationally optimize the layout of APs;

(3), define subspace information, position reference point information and path information of subspace on the map;

(4), determine the motion path of the user, and use the WLAN position investigation module to perform investigation correction and investigation test;

(5) Analyze the positioning accuracy of the system, and analyze the points exceeding the 3-meter error according to the positioning error vector diagram, and determine the cause of the error (such as insufficient number of APs, unreasonable AP layout, too few sampling points, too much spacing between sampling points large, the sampling time is too short, the environmental interference is too serious, etc.), and re-adjust the AP and re-sample the point until the positioning accuracy of the system meets the user's needs;

(6) Carry out real-time tracking experiments to verify the actual performance and reliability of the system.

The utility model has the following advantages:

The utility model can be based on the existing wireless local area network (802.11a/b/g), and the positioning method adopts a positioning algorithm based on received signal strength (RSS). Compared with other indoor positioning systems or equipment, it does not require Adding additional hardware devices for precise time synchronization and angle measurement makes full use of existing wireless network facilities, thus greatly reducing the hardware cost of the device.

The technical index and level of this indoor positioning equipment:

(1) Indoor positioning equipment based on 802.11 wireless local area network;

(2) In a typical office building environment, 80% of the positioning accuracy can be within 3 meters, and 90% of the positioning accuracy can be within 5 meters;

(3) The single positioning time of the system is short, generally less than 25 seconds.

Description of drawings

Below in conjunction with accompanying drawing and embodiment the utility model is described further.

Fig. 1 is a schematic diagram of the utility model

Fig. 2 is the schematic flow chart of positioning algorithm of the utility model

Figure 3 is a schematic diagram of the sampling path and AP layout of the experimental site

Fig. 4 is that the utility model adopts the comparison of two kinds of positioning algorithm experimental results of nearest neighbor method and probability method

Detailed ways

An indoor positioning device based on a wireless local area network, which includes a server monitoring module 1 that receives the location information sent by each client positioning module in real time and displays the user's location information on a map in real time, and defines subspaces, location reference points, and paths The location investigation module 2 of various information and test analysis, real-time tracking, wireless network card setting, etc., calculates the user's current location and sends the location information to the client positioning module 3 of the server monitoring system; the central database is 4. The server monitoring module, the location investigation module and the client positioning module communicate with each other through the wireless local area network. The wireless LAN is 802.11 wireless LAN. The location fingerprint database is stored centrally in the central database. see picture 1.

The client positioning module adopts a signal strength filtering method when reading signal strength data. The client location module adopts the user location filtering method when processing the user location information. When the location investigation module defines various information such as subspace, location reference point, path, etc., the path information of the indoor environment is used to rationally optimize the search subspace of the positioning algorithm, that is, the method of path tracking assistance. The operating system platform of the server monitoring module is Windows XP, the operating system platform of the Windows 2003Server location investigation module is Windows XP; the operating system platform of the client positioning module is WindowsXP/2000.

The flow chart of the positioning algorithm of the utility model is shown in Fig. 2, and is specifically described as follows:

First, the system reads the signal strength of each access point (AP) through the wireless network card driver, and then enters the sampling stage 5 and the positioning stage 6 respectively; in the sampling stage 5, the read signal strength is filtered by signal strength and feature information is extracted Save it in the fingerprint database afterwards, the content of the fingerprint database includes the position coordinates of each sampling point and the signal strength of each access point (AP) corresponding to the point; during the positioning stage 6, the read signal strength is filtered by signal strength , calculate the initial estimated position of the user by the nearest neighbor method, and then correct the estimated position through the user position filtering and path tracking auxiliary modules, and finally output the corrected user position coordinates for the upper application software to call.

Application experiment of the present utility model:

The experimental site for the indoor positioning of the utility model is the 11th floor of the Sakura Building, and the environment is a typical office building environment. In the experiments in this paper, we only conduct AP deployment experiments on a half-floor building of 405 square meters. The area has a length of 27 meters and a width of 15 meters, including 15 small rooms, a corridor, a toilet, a staircase and an elevator, for a typical office building environment.

We deployed a total of 8 access points (Access Point) in this area. The AP model is the most common TP-LINK TL-WA501G model in China. The AP layout should be asymmetrical or zigzag. , cannot be placed in a straight line, and the distance between APs should not be too small.

To obtain the signal strength of the wireless network card, we use Microsoft's Windows XP driver development kit (DDK) and the Wireless Research API (WRAPI) of the University of California. WRAPI is a software library developed in 2002. It is mainly used for mobile terminal equipment to query IEEE802.11 network related information, including MAC address, SSID and received signal strength. The deployment of the present utility model can generally be divided into following 6 steps:

Determine the venue for indoor positioning and draw a map file of the venue;

Determine the model and quantity of APs, and rationally optimize the layout of APs;

Define subspace information and location reference point information on the map;

Determine the user's movement path and use the location survey program to conduct survey corrections and survey tests;

Analyze the positioning accuracy of the system, and analyze the points exceeding the 3-meter error according to the positioning error vector diagram, determine the cause of the error, and readjust the AP and re-sample the point until the positioning accuracy of the system meets the user's needs;

Conduct real-time tracking experiments to verify the actual performance and reliability of the system.

The sampling path and AP layout of the experimental site are shown in Figure 3, and 7 is the access point AP.

The distance between each sampling point is about 1.5 meters, and the sampling time of each sampling point is 25 seconds, that is, the signal strength of each sampling point is read 50 times (the signal strength scanning frequency of this system is 2 times/second).

In addition, the subspace division standard is: each small room is a subspace, and the corridor outside the room, the corridor inside the room, the bathroom, the stairs and the elevator are separate subspaces, so there are 20 subspaces in this experimental environment. space.

The location algorithm of the utility model adopts known nearest neighbor method (kNN) and probability method (Probabilistic Method).

In this paper, the accuracy evaluation criteria of the indoor positioning system mainly refer to the Ekahau positioning system, which are the average error (AvgError), 90% positioning error (90% Error), subspace accuracy (Zone Accuracy) and accuracy within 3 meters ( Accuracy in3 meters).

The experimental results of the two localization algorithms are shown in Fig. 4. The experimental results show that the nearest neighbor method (kNN) can obtain 90% of the positioning accuracy within 2.7 meters, and the probabilistic method (Probabilistic Method) can obtain 90% of the positioning accuracy within 2.8 meters. The 90% positioning error of the two positioning algorithms is compared It is similar, but in terms of average error, subspace accuracy and accuracy within 3 meters, the nearest neighbor method is far more accurate than the probability method. The reason for this analysis is that the indoor signal strength is seriously interfered by the environment, and the sampling of the system The time is short (only 25 seconds for each sampling point), so that the Gaussian distribution characteristic of the signal intensity is not obvious.

That is, in a typical office building experimental environment, the utility model has achieved a positioning accuracy of 96% within 3 meters for the nearest neighbor method, and a positioning accuracy of 91% within 3 meters for the probabilistic method.

Claims (3)

1. based on the indoor positioning equipment of WLAN (wireless local area network), it is characterized in that, it comprises positional information that receives each client location module transmission in real time and the service end monitoring module that shows this user position information on map in real time, define the position inquiry module that subspace, position reference point, path and test analysis, real-time tracking, wireless network card are provided with, calculate user's current location and positional information is sent to the client location module of service end supervisory control system; Service end monitoring module, position inquiry module, client location module are carried out communication by WLAN (wireless local area network) each other.

2. the indoor positioning equipment based on WLAN (wireless local area network) according to claim 1 is characterized in that: described WLAN is 802.11 WLAN (wireless local area network).

3. the indoor positioning equipment based on WLAN (wireless local area network) according to claim 2 is characterized in that: the operating system platform of described service end monitoring module operation is Windows XP, Windows 2003 Server; The operating system platform of position inquiry module operation is Windows XP; The operating system platform of client location module operation is Windows XP/2000.

Images