CN106019209A - Indoor person device-free localization method based on radio tomography imaging - Google Patents

Abstract

The invention relates to a passive positioning method for indoor personnel based on radio frequency tomography. The method divides the positioning area into N uniform pixels, and arranges tags and readers on the periphery of the indoor area. It is assumed that there are M pixels between the readers and the tags. communication link, the shadow fading of the communication link between each pair of readers and tags is denoted as y, and the shadow fading on the pixel is denoted as x, and y has a linear relationship with x, that is, y=Wx+n, where W is the weight matrix, n is the additive Gaussian white noise, the personnel in the positioning area will block a part of the communication link, and the measured y value of this part of the communication link will fade a lot compared with the unblocked link. According to the above relationship, The shadow fading y of each communication link is assigned to each pixel according to the weight, and the x fading image of each pixel is reconstructed, and the position of the pixel with the most fading x is found to be the personnel position. The invention has the advantage of strong anti-interference ability.

Description

Passive positioning method of indoor personnel based on radio frequency tomography

technical field

The invention belongs to the research field of passive positioning of indoor personnel by using UHF RFID equipment, and specifically relates to radio frequency tomography technology.

Background technique

With the continuous development of wireless technology and mobile communication, people are more and more eager to obtain various target information in the environment at any time, any place, and in any way, and various positioning technologies have emerged, such as the well-known GPS. At present, outdoor positioning technology has developed relatively mature. Since indoors are different from outdoors, the environment is more complex, which makes the wireless signal transmission channel more complex, and those outdoor positioning technologies cannot be well applied to indoor positioning. Therefore, further research is needed on indoor positioning technologies. In recent years, indoor positioning technology has attracted the attention of many scholars at home and abroad, and has achieved many research results so far.

In the positioning process, the positioning target can participate in an active or passive manner. Active positioning often requires the positioning target to wear signal transceiver equipment. However, in the case of emergency response, intruder detection, smart home, etc., the positioning target is not equipped with any wireless transceiver equipment. At this time, another positioning technology - passive positioning (Device-Free localization, DFL) is needed. This positioning method is superior to the traditional passive positioning technology based on vision and infrared, and is not limited by weather, light, and line-of-sight requirements, and can perform target positioning in some special scenarios. To sum up, passive positioning has broad application prospects and research value.

Passive ultra-high frequency (UHF) RFID tags have the advantages of small size, low cost, simple deployment, no power supply, etc., and can have a longer communication distance and faster read and write speeds, and have been widely used in logistics and warehouse management. , robots, office buildings, parking lots and other environments. Tags and readers have become an important way to obtain target location information.

Radio Tomography Image (RTI) RTI technology has been widely used in passive positioning of wireless sensor networks, and is an imaging algorithm similar to CT scanning. Its basic idea is: deploy a wireless network in the area to be located, divide the area into multiple grids, here we call these grids pixels, and calculate the mutual communication between wireless network nodes by measuring the received signal strength RSS For link shadow fading, the shadow fading of each link is distributed to each pixel according to the weight, because the position of each pixel is known, and the position of the target can be determined by finding the pixel with the most shadow fading.

Contents of the invention

The invention provides a passive positioning method for indoor personnel with strong anti-interference ability. The technical scheme is as follows:

A passive positioning method for indoor personnel based on radio frequency tomography. This method divides the positioning area into N uniform pixels, and arranges UHF RFID tags and readers around the perimeter of the indoor area. It is assumed that there are M strips between the readers and the tags. For the communication link, the shadow fading of the communication link between each pair of readers and tags is denoted as y, and the shadow fading on the pixel is denoted as x, and y has a linear relationship with x, that is, y=Wx+n, where W is Weight matrix, n is additive Gaussian white noise, people in the positioning area will block a part of the communication link, and the measured y value of this part of the communication link will fade a lot compared with the link that is not blocked. According to the above relationship, each The shadow fading y of the communication link is assigned to each pixel according to the weight, and the x fading image of each pixel is reconstructed, and the position of the pixel with the most fading x is found to be the personnel position.

The specific steps of the passive positioning method for indoor personnel may be:

1) Select a rectangular area as the positioning area, deploy multiple UHF RFID tags evenly around the positioning area, and place a reader at the center of each side of the rectangle;

2) In view of the unique backscattering characteristics of UHF RFID, the ellipse weight model is adopted, and the shadow fading of the forward link and the backward link is considered, and the weights of the forward link and the rear weight link are calculated respectively. For pixels within the ellipse whose label is the focal point, the weight is set to the reciprocal of the root of the focal length inside the ellipse, and for pixels outside the ellipse, its weight is set to 0, and the adjustable parameters are used to adjust the weight range of the ellipse; set the mth communication The weights of the forward link and the backward link on the nth pixel of the link are w _m,n,forward and w _m,n,backward respectively; the weight matrix W is established, and the m and nth elements of W are w _{m ,n} =w _m,n,forward +w _m,n,backward ;

3) Measure the shadow fading y value of each communication link, use least squares and Tikhonov regularization to obtain the x value at each pixel, and perform image reconstruction on x: x=(W ^T W+αQ) ^-1 W ^T y, where α is a Tikhonov parameter and Q is a Tikhonov matrix;

4) Make the RFID device in the UHF frequency band perform frequency hopping communication on the 902MHz to 928MHz frequency band, perform image reconstruction on different frequency points respectively, and repeat step 3);

5) Analyzing the image reconstruction results obtained at each frequency point, judging the credibility of the positioning results, and dividing them into different levels β _f ;

6) According to the reliability level obtained in 5), the x obtained at each frequency point is defined as x _f , and the weighted average of each x _f can be obtained to obtain the final reconstructed image The formula is: reconstructed image The middle shadow fades the most The pixel position is the position of the person to be located.

Description of drawings

Fig. 1 is a flow chart of the present invention.

Figure 2 is a schematic diagram of a UHF RFID indoor passive positioning scenario.

detailed description

In order to further illustrate the present invention, a specific example is given below in conjunction with FIG. 1 and FIG. 2 . This example is limited to illustrate an implementation method of the present invention, and does not represent a limitation to the scope of coverage of the present invention.

The schematic diagram of the UHF RFID indoor passive positioning scene is shown in Figure 2. It is a 5m × 5m positioning area, and 16 tags and four readers are placed around it. Assuming that there are M signal links in total, the location area is divided into N uniform pixels, the shadow fading of all links is marked as y, and the shadow fading of pixels is marked as x. Since the shadow fading of each pixel is linearly added to the corresponding weight, that is, y=Wx+n, where W is the weight model and n is the additive Gaussian white noise. Obtaining y through measurement and deriving x in reverse, so as to achieve the purpose of passive positioning of personnel.

The specific method implementation process is described as follows:

1) Multiple passive tags are evenly deployed around the area to be located, and each reader is placed in the center of each side, dividing the area to be located into multiple pixels.

2) Calculating the radio frequency tomography weight model suitable for this scenario, considering the shadow fading of the forward link and the backward link respectively, and the weight models are w _m,n,forward and w _m,n,backward respectively. Wherein, the length of the m-th signal link is d _m , and the lengths from the center of the n-th pixel to the two endpoints of the m-th link are respectively expressed as d _1,m,n , d _2,m,n , which can be The tuning parameters λ _forward and λ _backward are respectively the difference between the length of the main axis of the ellipse and the inner focus distance d _m of the forward link and the backward link.

3) Measure the shadow fading value of the received signal strength of each link, and use least squares and Tikhonov regularization for image reconstruction. In the formula, α is a Tikhonov parameter, Q is a Tikhonov matrix, W=w _m,n,forward +w _m,n,backward .

$\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; Q</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}\mathrm{<span\; class="notranslate">\; the\; y</span>}$

4) Make the RFID device in the UHF frequency band perform frequency hopping communication in the 902MHz to 928MHz frequency band, perform image reconstruction on different frequencies respectively, and repeat step 3).

5) Analyze the image reconstruction results obtained at each frequency point, judge the reliability of the positioning results, and divide them into different levels β _f .

6) According to the reliability level obtained in 5), the weighted average of all image reconstruction results is carried out to obtain the final reconstructed image.

x＝Σβ _f x _f /Σβ _f

Claims (2)

1. A passive positioning method for indoor personnel based on radio frequency tomography, the method divides the positioning area into N uniform pixels, arranges UHF RFID tags and readers around the perimeter of the indoor area, and assumes that the reader and the tags share M communication links, the shadow fading of the communication link between each pair of readers and tags is recorded as y, and the shadow fading on the pixel is recorded as x, and y has a linear relationship with x, that is, y=Wx+n, where, W is the weight matrix, n is the additive Gaussian white noise, the personnel in the positioning area will block a part of the communication link, and the measured y value of this part of the communication link will fade a lot compared with the link that is not blocked. According to the above relationship The shadow fading y of each communication link is assigned to each pixel according to the weight, and the x fading image of each pixel is reconstructed, and the position of the pixel where x is found with the most shadow fading is the personnel position. 2. The indoor personnel passive positioning method according to claim 1, comprising the following steps: 1) Select a rectangular area as the positioning area, deploy multiple UHF RFID tags evenly around the positioning area, and place a reader at the center of each side of the rectangle; 2) In view of the unique backscattering characteristics of UHF RFID, the ellipse weight model is adopted, and the shadow fading of the forward link and the backward link is considered, and the weights of the forward link and the rear weight link are calculated respectively. For pixels within the ellipse whose label is the focal point, the weight is set to the reciprocal of the root of the focal length inside the ellipse, and for pixels outside the ellipse, its weight is set to 0, and the adjustable parameters are used to adjust the weight range of the ellipse; set the mth communication The weights of the forward link and the backward link on the nth pixel of the link are w _m,n,forward and w _m,n,backward respectively; the weight matrix W is established, and the m and nth elements of W are w _{m ,n} =w _m,n,forward +w _m,n,backward ; 3) Measure the shadow fading y value of each communication link, use least squares and Tikhonov regularization to obtain the x value at each pixel, and perform image reconstruction on x: x=(W ^T W+αQ) ^-1 W ^T y, where α is a Tikhonov parameter and Q is a Tikhonov matrix; 4) Make the RFID device in the UHF frequency band perform frequency hopping communication on the 902MHz to 928MHz frequency band, perform image reconstruction on different frequency points respectively, and repeat step 3); 5) Analyzing the image reconstruction results obtained at each frequency point, judging the credibility of the positioning results, and dividing them into different levels β _f ; 6) According to the reliability level obtained in 5), the x obtained at each frequency point is defined as x _f , and the weighted average of each x _f can be obtained to obtain the final reconstructed image The formula is: reconstructed image The middle shadow fades the most The pixel position is the position of the person to be located.