CN102625443B - Method and device for positioning terminal - Google Patents

Abstract

The invention discloses a method and a device for positioning a terminal, wherein the method comprises the steps of: for any cell, setting a first preset number of clustering terminals in a cell, acquiring position data of each clustering terminal: angle of arrival AOA, time of arrival TOA, signal to noise ratio SNR and the physical coordinate, and inputting the position data of each clustering terminal into a historical data base; for any terminal to be positioned, measuring the AOA, the TOA and the SNR of the terminal, based on the AOA, the TOA and the SNR of the terminal to be positioned, searching the historical database for a clustering terminal which matches best to the terminal to be positioned, and determining an initial estimate position of the terminal to be positioned based on the physical coordinate of the clustering terminal; calculating the final position of the terminal to be positioned based on the initial estimate position of the terminal to be positioned and the distance between the terminal to be positioned and other terminals to be positioned. According to the invention, only one base station is needed for positioning a terminal.

Description

Method of locating terminal and device

Technical field

The present invention relates to field of locating technology, be specifically related to method of locating terminal and device.

Background technology

Along with the development of mobile communication, the mobile location service in mobile communication system more and more receives publicity.There are at present 3 kinds of conventional localization methods:

1, the localization method based on time measurement, as: the time of advent (TOA, Time of Arrival) method, the time of advent poor (TDOA, Time Different Of Arrival) method.Main passing through the propagation time of detection electric wave from transmitted from transmitter to receiver, calculates the distance of the two, and then by calculating someway the estimated position of travelling carriage.

2, direction of arrival (AOA, Angle of Arrival) method: the AOA that detects travelling carriage emitting radio waves by base station receiver aerial array, this AOA can form a radially line from base station to travelling carriage, it is rhumb line, the AOA measured value that utilizes multiple base stations to provide, can draw many rhumb line, its intersection point is exactly the estimated position of travelling carriage.

3, auxiliary global satellite positioning system (AGPS, Assisted Global Positioning System) positioning mode: a kind ofly combine the technology that network base station information and GPS information position travelling carriage.Travelling carriage utilizes the GPS supplementary that network provides to receive GPS primary signal, obtain GPS pseudorange information by the demodulation to primary signal, network completes the processing to GPS information according to the supplementary of pseudorange information and other positioning equipment, and estimates the position of travelling carriage.

The shortcoming of prior art is as follows:

Localization method based on time measurement: its positioning precision is subject to the impact of timing precision, must be strict synchronous between base station, can not cause obvious impact to positioning result with the timing error of guaranteeing system itself; And terminal must minimumly be observed three base station signals and could locate.

Direction of arrival (AOA) method: location just can be realized in two base stations, but need to there be receiving antenna array base station, and precision is subject to channel effect larger, in building compact district location difficulty.

AGPS positioning mode: for positioning requirements at least receives the signal of four satellites, between intensive urban district high building, in building and can only see any area of satellite below 4, gps system is generally inoperative, cannot realize accurate location.

Summary of the invention

The invention provides method of locating terminal and device, only need a base station just can realize the location to terminal to reach.

Technical scheme of the present invention is achieved in that

A kind of method of locating terminal, the method comprises:

For arbitrary community, the first preset number cluster terminal is set in community, obtain the position data of each cluster terminal: angle of arrival AOA, the time of advent TOA, signal to noise ratio snr and physical coordinates, the position data of each cluster terminal is put into historical data base;

For arbitrary terminal to be measured, measure AOA, TOA and the SNR of this terminal to be measured, according to the AOA of this terminal to be measured, TOA and SNR, in historical data base, search the cluster terminal of mating most with this terminal to be measured, determine the initial estimated location of this terminal to be measured according to the physical coordinates of this cluster terminal;

According to the distance between the initial estimated location of terminal to be measured and this terminal to be measured and other terminal to be measured, the final position of calculating this terminal to be measured.

Described the first preset number cluster terminal that arranges in community, obtain the position data of each cluster terminal: angle of arrival AOA, the time of advent TOA, signal to noise ratio snr and physical coordinates, the position data of each cluster terminal is put into historical data base and comprises:

The first preset number initial clustering terminal is set in community, and the second preset number sample terminal is set;

According to the distance of sample terminal and cluster terminal, all sample terminals are carried out to cluster, cluster completes, obtain the first preset number new cluster terminal, calculate the position data of each new cluster terminal: AOA, TOA, SNR and physical coordinates, the position data of each new cluster terminal is put into historical data base.

Described according to the distance of sample terminal and cluster terminal, all sample terminals are carried out to cluster and comprise:

Initialization iterations is 0;

For each sample terminal, find the cluster terminal nearest with this sample terminal, by this sample terminal attaching in this cluster terminal, using all sample terminals that belong to same cluster terminal as a cluster group, using each cluster group's barycenter as a new cluster terminal, the physical coordinates of all new cluster terminal that calculating current iteration obtains and the mean square error of the physical coordinates of all cluster terminals that last iteration obtains; If mean square error is less than preset value, determine that cluster completes; Otherwise, return to execution described for each sample terminal, find the action of the cluster terminal nearest with this sample terminal.

Described the first preset number initial clustering terminal that arranges in community is:

Community is evenly divided into the first preset number sub regions, using the center of every sub regions as an initial clustering terminal position.

After the final position of described this terminal to be measured of calculating, further comprise:

When calculating behind the final position of this terminal to be measured, using this terminal to be measured as new sample terminal, carry out cluster with existing sample terminal, cluster completes, the position data with the each new cluster terminal that obtains: AOA, TOA, SNR and physical coordinates upgrade historical data base.

Described according to the AOA of this terminal to be measured, TOA and SNR, in historical data base, search the cluster terminal of mating most with this terminal to be measured and comprise:

According to the AOA of this terminal to be measured, TOA and SNR, in historical data base, search the cluster terminal with the Euclidean distance minimum of this terminal to be measured;

The described physical coordinates according to this cluster terminal determines that the initial estimated location of this terminal to be measured is:

Calculate the difference α of the AOA of described terminal to be measured and the AOA of described cluster terminal, by the physical coordinates rotation alpha of this cluster terminal, the initial estimated location using the physical coordinates obtaining after rotation as described terminal to be measured.

Described according to the distance between the initial estimated location of terminal to be measured and this terminal to be measured and other terminal to be measured, the final position of calculating this terminal to be measured further comprises before:

Calculate as follows the distance between described terminal to be measured and arbitrary other terminal to be measured:

Calculate β=AOA _i-AOA _k, wherein, AOA _ifor the AOA of described terminal to be measured, AOA _kfor the AOA of described arbitrary other terminal to be measured;

Calculate the distance of described terminal to be measured and base station calculate the distance of described arbitrary other terminal to be measured and base station wherein, for according to SNR _ithe terminal described to be measured calculating and the distance of base station, for according to TOA _ithe terminal described to be measured calculating and the distance of base station, for according to SNR _kdescribed arbitrary other terminal to be measured calculating and the distance of base station, for according to TOA _kdescribed arbitrary other terminal to be measured calculating and the distance of base station, SNR _ifor the SNR of described terminal to be measured, TOA _ifor the TOA of described terminal to be measured, SNR _kfor the SNR of described arbitrary other terminal to be measured, TOA _kfor the TOA of described arbitrary other terminal to be measured;

Calculate the distance d of described terminal to be measured and described arbitrary other terminal room to be measured _ik:

$d_{\mathrm{ik}}=\sqrt{d_i^2+d_k^2-2d_i{d}_k\·\cos \mathrm{\β}}.$

Described according to the distance between the initial estimated location of terminal to be measured and this terminal to be measured and other terminal to be measured, the final position of calculating this terminal to be measured comprises:

If the dimensional Gaussian that the position distribution of described terminal to be measured is obeyed centered by initial estimated location distributes, determine the initial position probability distribution of this terminal to be measured according to the initial estimated location of described terminal to be measured;

Initialization iterations is 0;

According to the distance between described terminal to be measured and arbitrary other terminal to be measured, calculate the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured in this iterative process;

According to the renewal value of the position probability distribution of other terminal to be measured to described terminal to be measured in the initial position probability distribution of described terminal to be measured and this iterative process, obtain the latest position probability distribution of described terminal to be measured;

Judge whether iterations is less than pre-determined number, if so, return to the action of carrying out the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured in this iterative process of described calculating, continue next iteration process; Otherwise, using the latest position probability distribution of terminal to be measured as its final position probability distribution, the final position of calculating described terminal to be measured according to this final position probability distribution.

In this iterative process of described calculating, the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured is:

Calculate

Wherein, the sequence number that k is arbitrary other terminal to be measured, the sequence number that i is described terminal to be measured, q is current iteration number of times, for the renewal value of the position probability distribution of arbitrary other terminal k to be measured to described terminal i to be measured in this iterative process, the physical coordinates that (s, t) is arbitrary other terminal k to be measured, for the position probability distribution of arbitrary other terminal k to be measured after upper once iteration;

$p\left(d_{\mathrm{ik}}\right|(x,y),(s,t))=\frac{1}{{2\mathrm{\π\σ}}^2}\exp \{-\frac{{(d_{\mathrm{ik}}-\sqrt{{(x-s)}^2+{(y-t)}^2})}^2}{2{\mathrm{\σ}}^2}\}$

The physical significance of this formula is: when described arbitrary other terminal k to be measured is in the time that physical location (s, t) is upper, and the probability of described terminal i to be measured on physical location (x, y); d _ikfor the distance between described arbitrary other terminal k to be measured and described terminal i to be measured;

Described according to the renewal value of the position probability distribution of other terminal to be measured to described terminal to be measured in the initial position probability distribution of described terminal to be measured and this iterative process, the latest position probability distribution that obtains described terminal to be measured is:

wherein, p _i(x, y) is the initial position probability distribution of described terminal to be measured;

Described final position of calculating described terminal to be measured according to this final position probability distribution for:

${\hat{x}}_i=\underset{x,y}{\mathrm{\Σ}}x\·P_i^{\′}(x,y)$

${\hat{y}}_i=\underset{x,y}{\mathrm{\Σ}}y\·P_i^{\′}(x,y)$

Wherein, the span of (x, y) is the overlay area of community.

A kind of terminal positioning device, this device comprises:

First module: for arbitrary community, the first preset number cluster terminal is set in community, obtain the position data of each cluster terminal: angle of arrival AOA, the time of advent TOA, signal to noise ratio snr and physical coordinates, the position data of each cluster terminal is put into historical data base;

Second unit: for arbitrary terminal to be measured, measure AOA, TOA and the SNR of this terminal to be measured, according to the AOA of this terminal to be measured, TOA and SNR, in historical data base, search the cluster terminal of mating most with this terminal to be measured, determine the initial estimated location of this terminal to be measured according to the physical coordinates of this cluster terminal, the initial estimated location of this terminal to be measured is sent to Unit the 3rd;

Unit the 3rd: according to the distance between the initial estimated location of terminal to be measured and this terminal to be measured and other terminal to be measured, the final position of calculating this terminal to be measured.

Described first module comprises:

The first module: the first preset number initial clustering terminal is set in community, and the second preset number sample terminal is set;

The second module: initialization iterations is 0; For each sample terminal, find the cluster terminal nearest with this sample terminal, by this sample terminal attaching in this cluster terminal, using all sample terminals that belong to same cluster terminal as a cluster group, using each cluster group's barycenter as a new cluster terminal, the physical coordinates of all new cluster terminal that calculating current iteration obtains and the mean square error of the physical coordinates of all cluster terminals that last iteration obtains; If mean square error is less than preset value, determine that cluster completes; Otherwise, return to execution described for each sample terminal, find the action of the cluster terminal nearest with this sample terminal.

Described Unit the 3rd is further used for, and when calculating behind the final position of this terminal to be measured, the position data by this terminal to be measured: AOA, TOA, SNR and physical coordinates send to first module;

Described first module receives after the position data of described terminal to be measured, using this terminal to be measured as new sample terminal, again all sample terminals are carried out to cluster, cluster completes, the position data with the each new cluster terminal that obtains: AOA, TOA, SNR and physical coordinates upgrade historical data base.

Described second unit comprises:

The 3rd module: for arbitrary terminal to be measured, measure AOA, TOA and the SNR of this terminal to be measured;

Four module: according to the AOA of this terminal to be measured, TOA and SNR, in historical data base, search the cluster terminal with the Euclidean distance minimum of this terminal to be measured, calculate the difference α of the AOA of described terminal to be measured and the AOA of described cluster terminal, by the physical coordinates rotation alpha of this cluster terminal, the initial estimated location using the physical coordinates obtaining after rotation as described terminal to be measured.

Described Unit the 3rd comprises:

The 5th module: calculate the distance of described terminal to be measured and arbitrary other terminal room to be measured, the each distance calculating is sent to the 6th module;

The 6th module: the position distribution of establishing described terminal to be measured is obeyed the dimensional Gaussian distribution centered by initial estimated location, determines the initial position probability distribution of this terminal to be measured according to the initial estimated location of described terminal to be measured; Initialization repeatly number of times is 0; According to the distance of described terminal to be measured and arbitrary other terminal room to be measured, calculate the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured in this iterative process; According to the renewal value of the position probability distribution of other terminal to be measured to described terminal to be measured in the initial position probability distribution of described terminal to be measured and this iterative process, obtain the latest position probability distribution of described terminal to be measured; Judge whether iterations is less than pre-determined number, if so, return to the action of carrying out the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured in this iterative process of described calculating, continue next iteration process; Otherwise, using the latest position probability distribution of terminal to be measured as its final position probability distribution, the final position of calculating described terminal to be measured according to this final position probability distribution.

Described the 5th module comprises:

The first submodule: calculate β=AOA _i-AOA _k, wherein, AOA _ifor the AOA of described terminal to be measured, AOA _kfor the AOA of described arbitrary other terminal to be measured, β is sent to the 3rd submodule;

The second submodule: the distance of calculating described terminal to be measured and base station calculate the distance of described arbitrary other terminal to be measured and base station wherein, for according to SNR _ithe terminal described to be measured calculating and the distance of base station, for according to TOA _ithe terminal described to be measured calculating and the distance of base station, for according to SNR _kdescribed arbitrary other terminal to be measured calculating and the distance of base station, for according to TOA _kdescribed arbitrary other terminal to be measured calculating and the distance of base station, by d _i, d _ksend to the 3rd submodule;

The 3rd submodule: the distance d that calculates described terminal to be measured and described arbitrary other terminal room to be measured _ik:

$d_{\mathrm{ik}}=\sqrt{d_i^2+d_k^2-2d_i{d}_k\·\cos \mathrm{\β}}.$

Described device is positioned on base station.

Compared with prior art, in the present invention, work in coordination with without multiple base stations, only need a base station measurement to obtain AOA, TOA and the SNR of terminal to be measured, just can realize the location to terminal; And, in one-time positioning process, can realize the location to multiple terminals simultaneously; And, do not need the supplemental characteristic outside existing protocol, just can realize the location to terminal, there is good compatibility and usability.

Brief description of the drawings

The method of locating terminal flow chart that Fig. 1 provides for the embodiment of the present invention;

The method flow diagram of setting up historical data base that Fig. 2 provides for the embodiment of the present invention;

Fig. 3 is for dividing schematic diagram in the community that the embodiment of the present invention provides;

The schematic diagram of the position of the cluster point that Fig. 4 provides for the embodiment of the present invention;

The method flow diagram to terminal positioning to be measured according to historical data base that Fig. 5 provides for the embodiment of the present invention;

The method flow diagram of the terminal i initial estimated location definite to be measured that Fig. 6 provides for the embodiment of the present invention;

Fig. 7 for the embodiment of the present invention provide according to the schematic diagram of location positioning terminal location to be measured of coupling cluster point;

Fig. 8 for the embodiment of the present invention provide according to the initial estimated location of terminal to be measured and terminal to be measured distance between any two, adopt sum-product algorithm to obtain the method flow diagram of the final estimated position of terminal to be measured;

The position distribution schematic diagram of the terminal to be measured that Fig. 9 provides for the embodiment of the present invention;

Figure 10 applies the embodiment of the present invention according to the initial estimated location of terminal to be measured and terminal to be measured distance between any two, adopts sum-product algorithm to obtain the schematic diagram of the final estimated position of terminal to be measured;

The composition diagram of the terminal positioning device that Figure 11 provides for the embodiment of the present invention.

Embodiment

The method of locating terminal flow chart that Fig. 1 provides for the embodiment of the present invention, as shown in Figure 1, its concrete steps are as follows:

Step 101: for arbitrary community, the first preset number sample point is set in community, according to the position of sample point, all sample points are carried out to cluster, obtain the second preset number cluster point, calculate the position data of each cluster point: AOA, TOA, SNR and physical coordinates, the position data of each cluster point is put into historical data base.

Each sample point is a mobile terminal that physical coordinates is known.

Step 102: for arbitrary terminal to be measured, AOA, TOA and the SNR of this terminal to be measured of base station measurement, according to the AOA of this terminal to be measured, TOA and SNR, in historical data base, search the cluster point mating most with this terminal to be measured, determine the initial estimated location of this terminal to be measured according to the physical coordinates of this cluster point.

Step 103: calculate the distance between terminal to be measured between two, according to the distance between the initial estimated location of terminal to be measured and this terminal to be measured and other terminal to be measured, the final position of calculating this terminal to be measured.

Step 104: using this terminal to be measured as new sample point, carry out cluster with existing sample point, cluster completes, the position data with each cluster point of obtaining: AOA, TOA, SNR and physical coordinates upgrade historical data base.

Below provide the embodiment of above-mentioned steps 101:

The method flow diagram of setting up historical data base that Fig. 2 provides for the embodiment of the present invention, as shown in Figure 2, its concrete steps are as follows:

Step 201: for arbitrary community, Jiang Gai community, base station is evenly divided into preset number: M(M>1) individual region, using the center in each region as an initial clustering point, base station initialization iterations l=0, the position of recording each initial clustering point and, a preset number sample point is set in Gai community, base station.

For example: for arbitrary community, shown in Ru Tu3Zhong community 0, first this microzonation is divided into 6 equilateral triangles, then each equilateral triangle is further subdivided into 4 equilateral triangles, obtain 24 equilateral triangles, again each equilateral triangle is divided into 4 equilateral triangles, like this, this community has comprised M=96 equilateral triangle altogether, using the center of each equilateral triangle in this M equilateral triangle as an initial clustering point, obtain altogether M initial clustering point, the positional representation of each cluster point is wherein, represent the initial horizontal coordinate of j cluster point, represent the initial ordinate of j cluster point.

Step 202: for the each sample point in this community, in M the cluster point that base station obtains after upper once iteration, find the cluster point nearest with this sample point, this sample point is belonged to this cluster point, using all sample points that belong to same cluster point as a cluster group, obtaining like this M cluster group, is D by each cluster group representation _j, j=0,1 ..., M-1.

Step 203: each cluster group's barycenter is calculated in base station, using each barycenter as a new cluster point, obtains M new cluster point, the position of recording M new cluster point so altogether

Arbitrary cluster group D _jthe computing formula of barycenter as follows:

${\text{s}}_j^{l+1}=\frac{1}{\mathrm{num}\left(D_j\right)}\underset{(x_i,y_i)\∈D_j}{\mathrm{\Σ}}{x}_i,t_j^{l+1}=\frac{1}{\mathrm{num}\left(D_j\right)}\underset{(x_i,y_i)\∈D_j}{\mathrm{\Σ}}{y}_i$

Wherein, l+1 is current iteration number of times, for the abscissa of j cluster point obtaining after current iteration, for the ordinate of j cluster point obtaining after current iteration, num (D _j) be the cluster group D obtaining after current iteration _jin the number of sample point, x _ifor cluster group D _jin the abscissa of i sample point, y _ifor cluster group D _jin the ordinate of i sample point.

Step 204: the mean square error E of the physical coordinates of M the cluster point obtaining after the physical coordinates of M new cluster point of base station calculating and last iteration:

$E=\frac{1}{M}\underset{j=1}{\overset{M}{\mathrm{\Σ}}}({(s_j^{l+1}-s_j^l)}^2+{(t_j^{l+1}-t_j^l)}^2)$

for the abscissa of j cluster point obtaining after last iteration, for the ordinate of j cluster point obtaining after last iteration

Step 205: base station judges that whether E is less than predetermined threshold value, if so, performs step 206; Otherwise, return to step 202, start new round iteration, the l+2 time iteration.

Step 206: base station determines that cluster completes, according to the position of the M obtaining after a dissemination channel model and current iteration cluster point, obtain the AOA of each cluster point, TOA and SNR, by the AOA of each cluster point, TOA, SNR and physical coordinates (x, y) are as the position data of this cluster point, with (AOA, TOA, SNR, (x, y)) form be stored in historical data base.

The schematic diagram of the position of the cluster point that Fig. 4 provides for the embodiment of the present invention, as shown in Figure 4, establishes base station location for (0,0), for arbitrary cluster point (x _r, y _r), r=0,1 ..., M-1, cluster point (x _r, y _r) AOA and TOA be:

${\mathrm{AOA}}_r=\arctan \frac{y_r}{x_r}\±\mathrm{\π}$

${\mathrm{TOA}}_r=\frac{\sqrt{{x_r}^2+{y_r}^2}}{c}$

Wherein, c is the light velocity.

The method flow diagram to terminal positioning to be measured according to historical data base that Fig. 5 provides for the embodiment of the present invention, as shown in Figure 5, its concrete steps are as follows:

Step 501: for the terminal i arbitrary to be measured in community, the pilot signal that send according to terminal i to be measured base station, records the measuring position data AOA of this terminal _i, TOA _iand SNR _i, in historical data base, search and (AOA _i, TOA _i, SNR _i) position data of the cluster point s of coupling, according to the physical coordinates (x of cluster point s _s, y _s) determine the initial estimated location (x of terminal i to be measured _ui, y _ui).

Step 502: the distance d between any two terminal i to be measured, k, according to AOA and the TOA of each terminal to be measured of measuring, is calculated in base station _ik.

If the measuring position data of terminal i to be measured, k are respectively (AOA _i, TOA _i, SNR _i), (AOA _k, TOA _k, SNR _k), the distance between terminal i to be measured, k can obtain by following process:

Step 01: β=AOA is calculated in base station _i-AOA _k.

Step 02: the distance of terminal i to be measured and base station, according to dissemination channel model, is calculated in base station $d_i=(d_{{\mathrm{SNR}}_i}+d_{{\mathrm{TOA}}_i})/2,$ Calculate the distance of terminal k to be measured and base station $d_k=(d_{{\mathrm{SNR}}_k}+d_{{\mathrm{TOA}}_k})/2.$

Wherein, for adopting prior art according to SNR _ithe terminal i to be measured calculating and the distance of base station, for adopting prior art according to TOA _ithe terminal i to be measured calculating and the distance of base station, for adopting prior art according to SNR _kthe terminal k to be measured calculating and the distance of base station, for adopting prior art according to TOA _kthe terminal k to be measured calculating and the distance of base station.

Step 03: base station utilizes the cosine law to calculate the distance d between terminal i to be measured, k _ik:

$d_{\mathrm{ik}}=\sqrt{d_i^2+d_k^2-2d_i{d}_k\·\cos \mathrm{\β}}$

Step 503: the dimensional Gaussian that the position distribution of establishing each terminal to be measured meets centered by the initial estimated location obtaining in step 201 distributes, distance between each terminal to be measured obtaining in integrating step 202, base station adopts and amasss iterative algorithm, obtains the final estimated position of each terminal to be measured.

Below provide the embodiment of above-mentioned steps 501:

The method flow diagram of the terminal i initial estimated location definite to be measured that Fig. 6 provides for the embodiment of the present invention, as shown in Figure 6, its concrete steps are as follows:

Step 601: establish (AOA _i, TOA _i, SNR _i) be the measuring position data of terminal i to be measured, (AOA _r, TOA _r, SNR _r, (x _r, y _r)) be the position data of r cluster point in historical data base, SNR, according to dissemination channel model, is passed through in base station _iand TOA _icalculate the distance of terminal i to be measured and base station now, the measuring position data of terminal i to be measured are expressed as

In the present embodiment, measure the close degree of the position data of each cluster point in the measuring position data of terminal i to be measured and historical data base with Euclidean distance, but due to the dimension difference of AOA, TOA, SNR, cannot directly calculate, therefore, unification is converted to each value apart from length here.

Step 602: for arbitrary cluster point r, base station is according to the physical coordinates (x of cluster point r _r, y _r), calculate the distance d of cluster point r and base station _r, the position data of cluster point r is expressed as (AOA _r, d _r, d _r).

Due to the position of cluster point r and the position of base station all known, therefore, the distance d of cluster point r and base station _rcan calculate.

Step 603: the AOA distance between terminal i to be measured and arbitrary cluster point r is calculated respectively in base station

$l_{\mathrm{AOA}}^r=|{\mathrm{AOA}}_i-A{\mathrm{OA}}_r|(((d_{{\mathrm{SNR}}_i}+d_{{\mathrm{TOA}}_i})/2+d_r)/2)$

Step 604: the Euclidean distance between the position data of the arbitrary cluster point r in measuring position data and the historical data base of terminal i to be measured is calculated respectively in base station, finds the cluster point s of Euclidean distance minimum.

The computing formula of the Euclidean distance between the measuring position data of terminal i to be measured and the position data of arbitrary cluster point r is as follows:

$\mathrm{\ϵ}={l_{\mathrm{AOA}}^r}^2+{(d_{{\mathrm{SNR}}_i}-d_r)}^2+{(d_{{\mathrm{TOA}}_i}-d_r)}^2$

Step 605: base station makes α=AOA _i-AOA _s, cluster is put to the physical coordinates rotation alpha of s, obtain new physical coordinates (x _ui, y _ui), be the initial estimated location of terminal i to be measured, wherein:

x _ui＝x _s·cosα-y _s·sinα

y _ui＝x _s·sinα+y _s·cosα

Fig. 7 has provided according to the schematic diagram of the location positioning terminal location to be measured of coupling cluster point.

In this step 605, the object of cluster being put to the physical coordinates rotation alpha of s is more accurate for the initial position estimation of terminal i to be measured is obtained.In actual applications, also can be directly by (x _s, y _s) as the initial estimated location of terminal i to be measured.

Below provide the embodiment of above-mentioned steps 503:

Fig. 8 for the embodiment of the present invention provide according to the initial estimated location of terminal to be measured and terminal to be measured distance between any two, adopt sum-product algorithm to obtain the method flow diagram of the final estimated position of terminal to be measured, as shown in Figure 8, its concrete steps are as follows:

Step 801: establish total N terminal to be measured in community, for arbitrary terminal i to be measured, establish the position distribution of terminal i to be measured and obey with initial estimated location (x _ui, y _ui) centered by dimensional Gaussian distribute, the initial position probability distribution P of terminal i to be measured _i(x, y) can be expressed as:

$P_i(x,y)=\frac{1}{{2\mathrm{\π\σ}}^2}\exp \{-\frac{{(x-x_{\mathrm{ui}})}^2+{(y-y_{\mathrm{ui}})}^2}{{2\mathrm{\σ}}^2}\}$

Wherein, the value of σ can be determined with experience as required.

For example: if the initial estimated location of terminal i to be measured is (5,8), the initial position probability distribution P of terminal i to be measured _i(x, y) as shown in Figure 9.

Step 802: base station initialization iterations q=0, for arbitrary terminal i to be measured, the position probability distribution of initialization terminal i to be measured, that is, and order

Step 803: the renewal value of the position probability distribution of arbitrary other terminal k to be measured to terminal i to be measured in this iterative process is calculated in base station

Wherein, the physical coordinates that (s, t) is terminal k to be measured, the span of (s, t) is identical with the coverage of community, for the position probability distribution after the q-1 time iteration of terminal k to be measured,

$p\left(d_{\mathrm{ik}}\right|(x,y),(s,t))=\frac{1}{{2\mathrm{\π\σ}}^2}\exp \{-\frac{{(d_{\mathrm{ik}}-\sqrt{{(x-s)}^2+{(y-t)}^2})}^2}{2{\mathrm{\σ}}^2}\}$

The physical significance of this formula is: when terminal k to be measured is in the time that physical location (s, t) is upper, and the probability of terminal i to be measured on physical location (x, y). represent: the renewal value of the position probability distribution with the position probability distribution after the q-1 time iteration of terminal k to be measured to terminal i to be measured.

Step 804: base station is according to the initial position probability distribution P of terminal i to be measured _ithe renewal value of the position probability distribution of other terminal to be measured to terminal i to be measured in (x, y) and this iterative process, obtains the latest position probability distribution of terminal i to be measured

$b_i^{\left(q\right)}(x,y)=p_i(x,y)\·\underset{k\≠i}{\mathrm{\Π}}{u}_{k->i}^{\left(q\right)}(x,y)$

Step 805: base station judges whether iterations q is less than pre-determined number Q, if so, makes q=q+1, returns to step 803, continues next iteration process; Otherwise, execution step 806.

The value of Q can rule of thumb be determined.

Step 806: base station determines that the final position probability distribution of terminal i to be measured is: the final estimated position of terminal i to be measured for:

$\begin{array}{c}{\hat{x}}_i=\underset{x,y}{\mathrm{\Σ}}x\·P_i^{\′}(x,y)\\ {\hat{y}}_i=\underset{x,y}{\mathrm{\Σ}}y\·P_i^{\′}(x,y)\end{array}$ Wherein, the span of (x, y) is identical with the coverage of community.

Obtaining the final estimated position of terminal i to be measured after, using terminal i to be measured as new sample point, adopt embodiment illustrated in fig. 2, with existing sample point re-start cluster with upgrade historical data base.

Figure 10 has provided one according to the initial estimated location of terminal to be measured and terminal to be measured distance between any two, and employing sum-product algorithm obtains the schematic diagram of the final estimated position of terminal to be measured, wherein, and P _i, i=0,1 ..., the initial position probability distribution that N-1 is terminal i to be measured, be the position probability distribution of terminal i to be measured after the q time iteration, be terminal k to be measured in the q time iterative process pass to terminal i to be measured renewal value.When each iteration, terminal room to be measured transmits such renewal value mutually between two, in conjunction with the initial position probability distribution P of each terminal oneself _i, obtain new position probability distribution more accurately.For example: for the terminal to be measured 1 in Figure 10, for the first time when iteration, other arbitrary terminal k's to be measured the renewal value that passes to terminal 1 to be measured is terminal 1 to be measured is like this position probability distribution after iteration for the first time not only comprise the initial position probability distribution of oneself, also comprised other terminal to be measured pass to its renewal value, thereby can obtain result more accurately, and then carry out iteration for the second time, due to the probability distribution of other terminal to be measured process iteration is also more accurate, passes to the renewal value of terminal 1 to be measured also more accurate, terminal 1 to be measured for the second time the result of iteration compare for the first time just more accurate.So, after iteration several times, each terminal to be measured can make full use of the information of other terminal to be measured, obtains result more accurately, thereby reaches the object of co-positioned.

The composition diagram of the terminal positioning device that Figure 11 provides for the embodiment of the present invention, as shown in figure 11, it mainly comprises: historical data is set up unit 111, initial position estimation unit 112, final position estimation unit 113 and historical data library unit 114, wherein:

Historical data is set up unit 111: for arbitrary community, the first preset number cluster terminal is set in community, obtain the position data of each cluster terminal: AOA, TOA, SNR and physical coordinates, the position data of each cluster terminal is put into historical data library unit 114.

Initial position estimation unit 112: for arbitrary terminal to be measured, measure AOA, TOA and the SNR of this terminal to be measured, according to the AOA of this terminal to be measured, TOA and SNR, in historical data library unit 114, search the cluster terminal of mating most with this terminal to be measured, determine the initial estimated location of this terminal to be measured according to the physical coordinates of this cluster terminal, the initial estimated location of this terminal to be measured is sent to final position estimation unit 113.

Final position estimation unit 113: the distance between the initial estimated location of the terminal to be measured of sending according to initial position estimation unit 112 and this terminal to be measured and other terminal to be measured, the final position of calculating this terminal to be measured.

In actual applications, historical data is set up unit 111 and can be comprised: the first module and the second module, wherein:

The first module: the first preset number initial clustering terminal is set in community, and the second preset number sample terminal is set.

The second module: initialization iterations is 0; For each sample terminal, find the cluster terminal nearest with this sample terminal, by this sample terminal attaching in this cluster terminal, using all sample terminals that belong to same cluster terminal as a cluster group, using each cluster group's barycenter as a new cluster terminal, the physical coordinates of all new cluster terminal that calculating current iteration obtains and the mean square error of the physical coordinates of all cluster terminals that last iteration obtains; If mean square error is less than preset value, determine that cluster completes; Otherwise, return to execution described for each sample terminal, find the action of the cluster terminal nearest with this sample terminal.

In actual applications, final position estimation unit 113 is further used for, and when calculating behind the final position of this terminal to be measured, the position data by this terminal to be measured: AOA, TOA, SNR and physical coordinates send to historical data to set up unit 111;

And, historical data is set up after the position data that unit 111 receives the terminal to be measured that final position estimation unit 113 sends, using this terminal to be measured as new sample terminal, again all sample terminals are carried out to cluster, cluster completes, the position data with the each new cluster terminal that obtains: AOA, TOA, SNR and physical coordinates upgrade the historical data in historical data library unit 114.

In actual applications, initial position estimation unit 112 can comprise: the 3rd module and four module, wherein:

The 3rd module: for arbitrary terminal to be measured, measure AOA, TOA and the SNR of this terminal to be measured, the AOA of this terminal to be measured, TOA and SNR are sent to four module.

Four module: AOA, TOA and the SNR of the terminal to be measured of sending according to the 3rd module, in historical data library unit 114, search the cluster terminal with the Euclidean distance minimum of this terminal to be measured, calculate the difference α of the AOA of described terminal to be measured and the AOA of described cluster terminal, by the physical coordinates rotation alpha of this cluster terminal, the initial estimated location using the physical coordinates obtaining after rotation as described terminal to be measured.

In actual applications, final position estimation unit 113 can comprise: the 5th module and the 6th module, wherein:

The 5th module: calculate the distance of terminal to be measured and arbitrary other terminal room to be measured, the each distance calculating is sent to the 6th module.

The 6th module: the initial estimated location that receives the terminal to be measured of sending initial position estimation unit 112, if the dimensional Gaussian that the position distribution of described terminal to be measured is obeyed centered by initial estimated location distributes, determine the initial position probability distribution of this terminal to be measured according to the initial estimated location of described terminal to be measured; Initialization repeatly number of times is 0; According to the distance of described terminal to be measured and arbitrary other terminal room to be measured, calculate the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured in this iterative process; According to the renewal value of the position probability distribution of other terminal to be measured to described terminal to be measured in the initial position probability distribution of described terminal to be measured and this iterative process, obtain the latest position probability distribution of described terminal to be measured; Judge whether iterations is less than pre-determined number, if so, return to the action of carrying out the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured in this iterative process of described calculating, continue next iteration process; Otherwise, using the latest position probability distribution of terminal to be measured as its final position probability distribution, the final position of calculating described terminal to be measured according to this final position probability distribution.

Wherein, the 5th module can comprise: the first submodule, the second submodule and the 3rd module, wherein:

The first submodule: calculate β=AOA _i-AOA _k, wherein, AOA _ifor the AOA of described terminal to be measured, AOA _kfor the AOA of described arbitrary other terminal to be measured, β is sent to the 3rd submodule.

The second submodule: the distance of calculating described terminal to be measured and base station calculate the distance of described arbitrary other terminal to be measured and base station wherein, for according to SNR _ithe terminal described to be measured calculating and the distance of base station, for according to TOA _ithe terminal described to be measured calculating and the distance of base station, for according to SNR _kdescribed arbitrary other terminal to be measured calculating and the distance of base station, for according to TOA _kdescribed arbitrary other terminal to be measured calculating and the distance of base station, by d _i, d _ksend to the 3rd submodule.

The 3rd submodule: the distance d that calculates described terminal to be measured and described arbitrary other terminal room to be measured _ik:

$d_{\mathrm{ik}}=\sqrt{d_i^2+d_k^2-2d_i{d}_k\·\cos \mathrm{\β}}.$

Terminal positioning device in the embodiment of the present invention can be positioned on base station.

The foregoing is only preferred embodiment of the present invention, in order to limit the present invention, within the spirit and principles in the present invention not all, any amendment of making, be equal to replacement, improvement etc., within all should being included in the scope of protection of the invention.

Claims (16)

1. a method of locating terminal, is characterized in that, the method comprises:

For arbitrary community, the first preset number cluster terminal is set in community, obtain the position data of each cluster terminal: angle of arrival AOA, the time of advent TOA, signal to noise ratio snr and physical coordinates, the position data of each cluster terminal is put into historical data base;

For arbitrary terminal to be measured, measure AOA, TOA and the SNR of this terminal to be measured, according to the AOA of this terminal to be measured, TOA and SNR, in historical data base, search the cluster terminal of mating most with this terminal to be measured, determine the initial estimated location of this terminal to be measured according to the physical coordinates of this cluster terminal;

According to the distance between the initial estimated location of terminal to be measured and this terminal to be measured and other terminal to be measured, the final position of calculating this terminal to be measured.

2. method according to claim 1, it is characterized in that, described the first preset number cluster terminal that arranges in community, obtain the position data of each cluster terminal: angle of arrival AOA, the time of advent TOA, signal to noise ratio snr and physical coordinates, the position data of each cluster terminal is put into historical data base and comprises:

The first preset number initial clustering terminal is set in community, and the second preset number sample terminal is set;

According to the distance of sample terminal and cluster terminal, all sample terminals are carried out to cluster, cluster completes, obtain the first preset number new cluster terminal, calculate the position data of each new cluster terminal: AOA, TOA, SNR and physical coordinates, the position data of each new cluster terminal is put into historical data base.

3. method according to claim 2, is characterized in that, described according to the distance of sample terminal and cluster terminal, all sample terminals is carried out to cluster and comprise:

Initialization iterations is 0;

For each sample terminal, find the cluster terminal nearest with this sample terminal, by this sample terminal attaching in this cluster terminal, using all sample terminals that belong to same cluster terminal as a cluster group, using each cluster group's barycenter as a new cluster terminal, the physical coordinates of all new cluster terminal that calculating current iteration obtains and the mean square error of the physical coordinates of all cluster terminals that last iteration obtains; If mean square error is less than preset value, determine that cluster completes; Otherwise, return to execution described for each sample terminal, find the action of the cluster terminal nearest with this sample terminal.

4. method according to claim 2, is characterized in that, described the first preset number initial clustering terminal that arranges in community is:

Community is evenly divided into the first preset number sub regions, using the center of every sub regions as an initial clustering terminal position.

5. method according to claim 2, is characterized in that, after the final position of described this terminal to be measured of calculating, further comprises:

When calculating behind the final position of this terminal to be measured, using this terminal to be measured as new sample terminal, carry out cluster with existing sample terminal, cluster completes, position data with the each new cluster terminal that obtains is upgraded historical data base, and the position data of described new cluster terminal comprises: AOA, TOA, SNR and physical coordinates.

6. method according to claim 1, is characterized in that, described according to the AOA of this terminal to be measured, TOA and SNR, searches the cluster terminal of mating most with this terminal to be measured and comprise in historical data base:

According to the AOA of this terminal to be measured, TOA and SNR, in historical data base, search the cluster terminal with the Euclidean distance minimum of this terminal to be measured;

The described physical coordinates according to this cluster terminal determines that the initial estimated location of this terminal to be measured is:

Calculate the difference α of the AOA of described terminal to be measured and the AOA of described cluster terminal, by the physical coordinates rotation alpha of this cluster terminal, the initial estimated location using the physical coordinates obtaining after rotation as described terminal to be measured.

7. method according to claim 1, is characterized in that, described according to the distance between the initial estimated location of terminal to be measured and this terminal to be measured and other terminal to be measured, and the final position of calculating this terminal to be measured further comprises before:

Calculate as follows the distance between described terminal to be measured and arbitrary other terminal to be measured:

Calculate β=AOA _i-AOA _k, wherein, AOA _ifor the AOA of described terminal to be measured, AOA _kfor the AOA of described arbitrary other terminal to be measured;

Calculate the distance of described terminal to be measured and base station calculate the distance of described arbitrary other terminal to be measured and base station wherein, for according to SNR _ithe terminal described to be measured calculating and the distance of base station, for according to TOA _ithe terminal described to be measured calculating and the distance of base station, for according to SNR _kdescribed arbitrary other terminal to be measured calculating and the distance of base station, for according to TOA _kdescribed arbitrary other terminal to be measured calculating and the distance of base station, SNR _ifor the SNR of described terminal to be measured, TOA _ifor the TOA of described terminal to be measured, SNR _kfor the SNR of described arbitrary other terminal to be measured, TOA _kfor the TOA of described arbitrary other terminal to be measured;

Calculate the distance d of described terminal to be measured and described arbitrary other terminal room to be measured _ik:

$d_{\mathrm{ik}}=\sqrt{d_i^2+d_k^2-2d_i{d}_k\·\cos \mathrm{\β}}.$

8. method according to claim 1, is characterized in that, described according to the distance between the initial estimated location of terminal to be measured and this terminal to be measured and other terminal to be measured, the final position of calculating this terminal to be measured comprises:

If the dimensional Gaussian that the position distribution of described terminal to be measured is obeyed centered by initial estimated location distributes, determine the initial position probability distribution of this terminal to be measured according to the initial estimated location of described terminal to be measured;

Initialization iterations is 0;

According to the distance between described terminal to be measured and arbitrary other terminal to be measured, calculate the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured in this iterative process;

According to the renewal value of the position probability distribution of other terminal to be measured to described terminal to be measured in the initial position probability distribution of described terminal to be measured and this iterative process, obtain the latest position probability distribution of described terminal to be measured;

Judge whether iterations is less than pre-determined number, if so, return to the action of carrying out the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured in this iterative process of described calculating, continue next iteration process; Otherwise, using the latest position probability distribution of terminal to be measured as its final position probability distribution, the final position of calculating described terminal to be measured according to this final position probability distribution.

9. method according to claim 8, is characterized in that, in this iterative process of described calculating, the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured is:

Calculate

Wherein, the sequence number that k is arbitrary other terminal to be measured, the sequence number that i is described terminal to be measured, q is current iteration number of times, for the renewal value of the position probability distribution of arbitrary other terminal k to be measured to described terminal i to be measured in this iterative process, the physical coordinates that (s, t) is arbitrary other terminal k to be measured, for the position probability distribution of arbitrary other terminal k to be measured after upper once iteration;

$p\left(d_{\mathrm{ik}}\right|(x,y),(s,t))=\frac{1}{{2\mathrm{\π\σ}}^2}\exp \{-\frac{{(d_{\mathrm{ik}}-\sqrt{{(x-s)}^2+{(y-t)}^2})}^2}{2{\mathrm{\σ}}^2}\}$

The physical significance of this formula is: when described arbitrary other terminal k to be measured is in the time that physical location (s, t) is upper, and the probability of described terminal i to be measured on physical location (x, y); d _ikfor the distance between described arbitrary other terminal k to be measured and described terminal i to be measured;

Described according to the renewal value of the position probability distribution of other terminal to be measured to described terminal to be measured in the initial position probability distribution of described terminal to be measured and this iterative process, the latest position probability distribution that obtains described terminal to be measured is:

wherein, p _i(x, y) is the initial position probability distribution of described terminal to be measured;

Described final position of calculating described terminal to be measured according to this final position probability distribution for:

${\hat{x}}_i=\underset{x,y}{\mathrm{\Σ}}x\·P_i^{\′}(x,y)$

${\hat{y}}_i=\underset{x,y}{\mathrm{\Σ}}y\·P_i^{\′}(x,y)$

Wherein, the span of (x, y) is the overlay area of community.

10. a terminal positioning device, is characterized in that, this device comprises:

Historical data is set up unit: for arbitrary community, the first preset number cluster terminal is set in community, obtain the position data of each cluster terminal: angle of arrival AOA, the time of advent TOA, signal to noise ratio snr and physical coordinates, the position data of each cluster terminal is put into historical data base;

Initial position estimation unit: for arbitrary terminal to be measured, measure AOA, TOA and the SNR of this terminal to be measured, according to the AOA of this terminal to be measured, TOA and SNR, in historical data base, search the cluster terminal of mating most with this terminal to be measured, determine the initial estimated location of this terminal to be measured according to the physical coordinates of this cluster terminal, the initial estimated location of this terminal to be measured is sent to final position estimation unit;

Final position estimation unit: according to the distance between the initial estimated location of terminal to be measured and this terminal to be measured and other terminal to be measured, the final position of calculating this terminal to be measured.

11. devices according to claim 10, is characterized in that, described historical data is set up unit and comprised:

The first module: the first preset number initial clustering terminal is set in community, and the second preset number sample terminal is set;

The second module: initialization iterations is 0; For each sample terminal, find the cluster terminal nearest with this sample terminal, by this sample terminal attaching in this cluster terminal, using all sample terminals that belong to same cluster terminal as a cluster group, using each cluster group's barycenter as a new cluster terminal, the physical coordinates of all new cluster terminal that calculating current iteration obtains and the mean square error of the physical coordinates of all cluster terminals that last iteration obtains; If mean square error is less than preset value, determine that cluster completes; Otherwise, return to execution described for each sample terminal, find the action of the cluster terminal nearest with this sample terminal.

12. devices according to claim 11, it is characterized in that, described final position estimation unit is further used for, when calculating behind the final position of this terminal to be measured, send to historical data to set up unit the position data of this terminal to be measured, the position data of described terminal to be measured comprises: AOA, TOA, SNR and physical coordinates;

Described historical data is set up after the position data that unit receives described terminal to be measured, using this terminal to be measured as new sample terminal, again all sample terminals are carried out to cluster, cluster completes, position data with the each new cluster terminal that obtains is upgraded historical data base, and the position data of described new cluster terminal comprises: AOA, TOA, SNR and physical coordinates.

13. devices according to claim 10, is characterized in that, described initial position estimation unit comprises:

The 3rd module: for arbitrary terminal to be measured, measure AOA, TOA and the SNR of this terminal to be measured;

Four module: according to the AOA of this terminal to be measured, TOA and SNR, in historical data base, search the cluster terminal with the Euclidean distance minimum of this terminal to be measured, calculate the difference α of the AOA of described terminal to be measured and the AOA of described cluster terminal, by the physical coordinates rotation alpha of this cluster terminal, the initial estimated location using the physical coordinates obtaining after rotation as described terminal to be measured.

14. devices according to claim 10, is characterized in that, described final position estimation unit comprises:

The 5th module: calculate the distance of described terminal to be measured and arbitrary other terminal room to be measured, the each distance calculating is sent to the 6th module;

The 6th module: the position distribution of establishing described terminal to be measured is obeyed the dimensional Gaussian distribution centered by initial estimated location, determines the initial position probability distribution of this terminal to be measured according to the initial estimated location of described terminal to be measured; Initialization repeatly number of times is 0; According to the distance of described terminal to be measured and arbitrary other terminal room to be measured, calculate the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured in this iterative process; According to the renewal value of the position probability distribution of other terminal to be measured to described terminal to be measured in the initial position probability distribution of described terminal to be measured and this iterative process, obtain the latest position probability distribution of described terminal to be measured; Judge whether iterations is less than pre-determined number, if so, return to the action of carrying out the renewal value of the position probability distribution of arbitrary other terminal to be measured to described terminal to be measured in this iterative process of described calculating, continue next iteration process; Otherwise, using the latest position probability distribution of terminal to be measured as its final position probability distribution, the final position of calculating described terminal to be measured according to this final position probability distribution.

15. devices according to claim 14, is characterized in that, described the 5th module comprises:

The first submodule: calculate β=AOA _i-AOA _k, wherein, AOA _ifor the AOA of described terminal to be measured, AOA _kfor the AOA of described arbitrary other terminal to be measured, β is sent to the 3rd submodule;

The second submodule: the distance of calculating described terminal to be measured and base station calculate the distance of described arbitrary other terminal to be measured and base station wherein, for according to SNR _ithe terminal described to be measured calculating and the distance of base station, for according to TOA _ithe terminal described to be measured calculating and the distance of base station, for according to SNR _kdescribed arbitrary other terminal to be measured calculating and the distance of base station, for according to TOA _kdescribed arbitrary other terminal to be measured calculating and the distance of base station, by d _i, d _ksend to the 3rd submodule;

The 3rd submodule: the distance d that calculates described terminal to be measured and described arbitrary other terminal room to be measured _ik:

$d_{\mathrm{ik}}=\sqrt{d_i^2+d_k^2-2d_i{d}_k\·\cos \mathrm{\β}}.$

16. according to claim 10 to 15 arbitrary described devices, it is characterized in that, described device is positioned on base station.