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

Abstract

The invention relates to an indoor person device-free localization method based on radio tomography imaging. According to the method, a localization region is divided into N uniform pixels, and tags and readers are arranged at the periphery of an indoor region. The method assumes that there are M communication links between the readers and the tags, the shadow fading of the communication link between each pair of reader and tag is recorded as y and the shadow fading on the pixels is recorded as x, there is a linear relationship between y and x, namely, y=Wx+n, wherein W is a weight matrix, and n is additive Gaussian white noise. A person in the localization region blocks part of the communication links, and the measured y values of the communication links fade a lot compared with the links not blocked. Based on the relationship, the shadow fading y of each communication link is assigned to the corresponding pixel according to the weight, an x fading image of each pixel is reconstructed, and the position of the pixel of which x fades most is found, which is the position of the person. The method of the invention has the advantage of strong anti-jamming capability.

Description

Indoor occupant passive type localization method based on radio frequency tomography

Technical field

The invention belongs to the research field utilizing UHF RFID device to carry out indoor occupant passive type location, tool Body relates to radio frequency chromatography imaging technique.

Background technology

Along with wireless technology and the development of mobile communication, people increasingly thirst at any time, arbitrarily Various target informations in place, in any way acquisition environment, have expedited the emergence of out various location technology therewith, Than GPS as is well known.At present, outdoor positioning technology has developed to obtain comparative maturity.Due to, room Inside being different from outdoor, environment is more complicated so that transmission of wireless signals channel is more complicated, and those are outdoor fixed Position technology can not well be applied to indoor positioning, accordingly, it would be desirable to research should can be used for determining of indoor further Position technology.Indoor positioning technologies by domestic and international vast focus of attention, the most also obtained numerous research in recent years Achievement.

During location, location target can by actively or passive in the way of participate in.Actively Formula location often requires that location target wears signal sending and receiving equipment.But, in emergency reaction, detect intruder, When Smart Home, location target is not equipped with any wireless transmitting-receiving equipments, needs another this time Plant location technology--passive type location (Device-Free localization, DFL).This location mode It is better than traditional view-based access control model, infrared passive location technology, is not limited by weather, light, line-of-sight requirement etc. System, can carry out target location under some special screnes.In sum, passive type location has wide answering With prospect and researching value.

Passive ultra-high frequency (UHF) RFID label tag has that volume is little, low cost, deployment are simple, without power supply Etc. advantage, and can have longer communication distance and read or write speed faster, be widely applied to In the environment such as logistics, warehousing management, robot, office building, parking lot.Label and reader have become as The important channel that target position information obtains.

Radio frequency tomography (Radio Tomography Image, RTI) RTI technology has been widely used for Wireless sensor network passive type positions, and is a kind of similar CT scan imaging algorithm.Its basic thought is: Arranging wireless network in region to be positioned, this region is divided into multiple grid, we claim these nets here Lattice are pixel, by measuring received signal strength RSS, calculate each wireless network node intercommunication chain The shadow fading on road, by the shadow fading of each of the links by weight distribution in each pixel, because each picture The position of element is it is known that the position of target can be determined by finding the most pixel of shadow fading.

Summary of the invention

The present invention provides the indoor occupant passive type localization method that a kind of capacity of resisting disturbance is strong.Technical scheme is as follows:

A kind of indoor occupant passive type localization method based on radio frequency tomography, region, location is drawn by the method It is divided into N number of uniform pixel, arranges UHF RFID label tag and reader in the periphery of room area, if reading Total M communication link, the shadow fading of the communication link between every pair of reader and label between device and label Being designated as y, the shadow fading in pixel is designated as x, y and x and has linear relationship, i.e. y=Wx+n, wherein, and W For weight matrix, n is additive white Gaussian noise, and personnel are in region, location, can block a part of communication chain Road, this section communication link is measured y value and can be declined a lot compared to the link not being blocked, according to relation above Each communication link shadow fading y is assigned to each pixel by weight, reconstructs the x rejection image of each pixel, find The position of pixel residing for most x that declines is personnel positions.

Described indoor occupant passive type localization method, concrete steps may is that

1) selected rectangular area is as region, location, and in region, location, surrounding uniformly disposes multiple UHF RFID mark Sign, respectively place a reader in the center on each bar limit of rectangle；

2) for the distinctive backscattered feature of UHF RFID, oval weight model is used, it is considered to forward chaining The shadow fading of road and backward link, calculates the weight of forward link and rear power link respectively, be in reader and Label is the pixel in the range of the ellipse of focus, and it is reciprocal, outside ellipse that weights are set to the evolution of focal length in ellipse Pixel, its weights are set to 0, and utilize the oval proportion range of adjustable parameter regulation；If m communication link exists Forward link and the weight of backward link in nth pixel are respectively w_m,n,forward、w_m,n,backward；Set up weight Matrix W, the m of W, n element w_m,n=w_m,n,forward+w_m,n,backward；

3) measurement obtains the shadow fading y value of every communication link, utilizes least square and Tikhonov regularization Obtain the x value at each pixel, and x is carried out image reconstruction: x=(W^TW+αQ)^-1W^TY, in formula, α is Tikhonov parameter, Q is Tikhonov matrix；

4) RFID device making uhf band carries out frequency hopping communications in 902MHz to 928MHz frequency range, Carry out image reconstruction on different frequent points respectively, repeat step 3)；

5) analyze the image reconstruction result obtained at each frequency, it is judged that the credibility of its positioning result, and divide Different brackets β_f；

6) according to 5) in the confidence level that obtains, the x that each frequency obtains is defined as x_f, to each x_fAdd Weight average, can obtain final reconstruct imageFormula is:Reconstruct imageIn Shadow fading is mostResiding location of pixels is personnel's location to be positioned.

Accompanying drawing explanation

Fig. 1 is FB(flow block) of the present invention.

Fig. 2 is that UHF RFID indoor occupant passive type positions scene schematic diagram.

Detailed description of the invention

In order to further illustrate the present invention, provide an instantiation below in conjunction with Fig. 1 and Fig. 2.This example is only It is limited to a kind of implementation that the present invention is described, does not represent the restriction to coverage of the present invention.

UHF RFID indoor occupant passive type location scene schematic diagram is as in figure 2 it is shown, be determining of a 5m × 5m Region, position, places 16 labels and four readers the most respectively.Assume total M bars link, location Region is divided into N number of uniform pixel, and the shadow fading of all links is designated as y, the shadow fading note in pixel For x.Due to shadow fading and the respective weights linear, additive, i.e. y=Wx+n of each pixel, wherein, W is Weight model, n is additive white Gaussian noise.Obtained by measurement that y is counter releases x, reach personnel's Passive Positioning Purpose.

It is as follows that concrete method realizes process prescription:

1) uniformly disposing multiple passive label in area to be targeted surrounding, each reader is individually positioned in each edge Center, area to be targeted is divided into multiple pixel.

2) calculating is applicable to the radio frequency tomography weight model of this scene, considers forward link and backward chain respectively The shadow fading on road, its weight model is respectively w_m,n,forward、w_m,n,backward.Wherein, m bars link A length of d_m, the length of two end points at the center of nth pixel to m article of link is expressed as d_1,m,n, d_2,m,n, adjustable parameter λ_forward、λ_backwardBe respectively the oval main axis length of forward link and backward link with Interior focal length d_mDifference.

3) measurement draws the shadow fading value of each of the links received signal strength, utilizes least square and Ji Hong promise husband (Tikhonov) regularization carries out image reconstruction.In formula, α is Tikhonov parameter, and Q is Tikhonov matrix, W=w_m,n,forward+w_m,n,backward。

$\hat{x}={(W^{T}W+\mathrm{\α}Q)}^{-1}{W}^{T}y$

4) RFID device making uhf band carries out frequency hopping communications in 902MHz to 928MHz frequency range, Carry out image reconstruction the most respectively, repeat step 3).

5) analyze the image reconstruction result obtained at each frequency, it is judged that the credibility of its positioning result, and divide Different brackets β_f。

6) according to 5) in the confidence level that obtains, be weighted averagely, obtaining to all image reconstruction results Whole reconstruct image.

X=Σ β_fx_f/Σβ_f

Claims (2)

1. an indoor occupant passive type localization method based on radio frequency tomography, region, location is divided into N number of by the method Uniform pixel, arranges UHF RFID label tag and reader in the periphery of room area, if total M between reader and label Communication link, the shadow fading of the communication link between every pair of reader and label is designated as y, and the shadow fading in pixel is designated as X, y and x have linear relationship, i.e. y=Wx+n, and wherein, W is weight matrix, and n is additive white Gaussian noise, personnel Being in region, location, can block a part of communication link, this section communication link measures y value compared to the link not being blocked Can decline a lot, according to relation above, each communication link shadow fading y is assigned to each pixel by weight, reconstruct the x of each pixel Rejection image, finds the position of pixel residing for the most x of shadow fading to be personnel positions.

Indoor occupant passive type localization method the most according to claim 1, it is characterised in that comprise the following steps:

1) selected rectangular area is as region, location, and in region, location, surrounding uniformly disposes multiple UHF RFID label tag, at square A reader is respectively placed in the center on each bar limit of shape；

2) for the distinctive backscattered feature of UHF RFID, oval weight model is used, it is considered to forward link and backward chain The shadow fading on road, calculates the weight of forward link and rear power link respectively, is in reader and oval scope that label is focus In pixel, it is reciprocal that weights are set to the evolution of focal length in ellipse, and the pixel outside ellipse, its weights are set to 0, and utilize adjustable The oval proportion range of parameter regulation；If m communication link forward link in nth pixel and the rear weight to link are divided Wei w_m,n,forward、w_m,n,backward；Set up weight matrix W, the m of W, n element w_m,n=w_m,n,forward+w_m,n,backward；

3) measurement obtains the shadow fading y value of every communication link, utilizes least square and Tikhonov regularization to obtain respectively X value at pixel, and x is carried out image reconstruction: x=(W^TW+αQ)^-1W^TY, in formula, α is Tikhonov parameter, Q For Tikhonov matrix；

4) RFID device making uhf band carries out frequency hopping communications, at different frequent points in 902MHz to 928MHz frequency range The upper image reconstruction that carries out respectively, repeats step 3)；

5) analyze the image reconstruction result obtained at each frequency, it is judged that the credibility of its positioning result, and divide different brackets β_f；

6) according to 5) in the confidence level that obtains, the x that each frequency obtains is defined as x_f, to each x_fIt is weighted average, can To obtain final reconstruct imageFormula is:Reconstruct imageMiddle shadow fading is mostResiding Location of pixels is personnel's location to be positioned.