CN103491628A - NLOS transmission environment wireless locating method based on TPOAs - Google Patents

Abstract

Disclosed is an NLOS transmission environment wireless locating method based on TPOAs. The locating method comprises the following steps that (1) signals sent by an MS are received by a plurality of BSes, TOA information in the signals is extracted, and the information is collected to the main locating base station; (2) a measurement value of the main base station is used as a reference, and TOA measured by the other base stations are multiplied by the reference value to obtain the TPOAs; (3) according to the locating geometrical principle, a locating equation set with the TPOAs being parameters is established; (4) linearization processing is conducted on the locating equation set; (5) the equation set is solved by the utilization of the LS method, and locating is conducted. The NLOS transmission environment wireless locating method based on the TPOAs can effectively improve the locating precision.

Description

A kind of NLOS transmission environment wireless location method based on TPOA

Technical field

The present invention relates to the wireless location technology field, especially a kind of NLOS transmission environment wireless location method.

Background technology

Wireless location refers to a kind of technology that the parameters such as angle and distance that receive in signal are estimated mobile terminal locations that is included in of utilizing.In recent years, demand due to economic development and people's life, this technology has obtained application widely, it can provide the business that comprises distress call, travel information service, vehicle management etc., also be applied in the Fare Collection System and intelligent transportation system of position-based information, be the important component part of Internet of Things simultaneously.

In actual wireless transmission environment, due to a large amount of existence of barrier, so signal can't be along linear transmission in from transmitting terminal to received this period, and it often needs can arrive receiving terminal through emission and diffraction.This makes receiving terminal adjust the distance and the decline of the information estimator accuracy such as angle, thereby has significantly reduced the precision of wireless location algorithm.Accordingly, in the practical application of wireless location technology, reduce and even to reduce the error that non line of sight (NLOS, non-line-of-sight) transmission brings and be necessary.According to Motorola and Ericsson, the field survey at GSM networking is found, the NLOS error has along with travelling carriage (MS, mobile station) and base station or base station (BS, base station) increase of air line distance between and the trend that rises, this has just more aggravated the impact on traditional location algorithm precision.

Summary of the invention

In order to overcome the poor deficiency of precision of existing NLOS transmission environment wireless location method, the invention provides a kind of NLOS transmission environment wireless location method based on the time of advent long-pending (TPOA, time product ofarrival) of effective lifting positioning precision.

The technical solution adopted for the present invention to solve the technical problems is:

A kind of NLOS transmission environment wireless location method based on TPOA, described localization method comprises the following steps:

1) receive by a plurality of base station BSs the signal that mobile station MS sends, and extract (TOA, the time of arrival) information time of advent in signal, and these information are pooled in the dominant base of location;

2) take the measured value of dominant base is reference, and the TOA of other base station measurements is multiplied by reference value, is transformed into TPOA;

3), according to the location geometrical principle, build and take the positioning equation group that TPOA is parameter;

4) the positioning equation group is carried out to linearization process;

5) utilize least square (LS, least squares) Algorithm for Solving equation group, and position.

Further, described step 2) in, if utilize i base station BS _imeasure TOA _i, dominant base is made as the 1st base station, and the product of it and the 1st base station measurements is exactly TPOA so _i1, namely

TPOA _i1＝TOA _i*TOA ₁ (1)；

Distance between mobile station MS and i base station BS is write as:

$r_i=\sqrt{{(x-x_i)}^2+{(y-y_i)}^2}={\mathrm{TOA}}_i\×c---\left(2\right)$

Wherein (x, y) and (x _i, y _i) respectively travelling carriage be the coordinate of MS and i BS, c is propagation velocity in the electromagnetic wave air;

According to above-mentioned formula, TPOA is transformed into to distance long-pending, if make dp _i1for the product between range finding, and do not lose a characteristic of stock, suppose BS ₁position be positioned at the origin of coordinates, so

$d{p}_{i1}^2=(x^2+y^2)({(x-x_i)}^2+{(y-y_i)}^2)---\left(3\right)$

Further, described step 4) in, (3) are carried out to mathematic(al) manipulation

$d{p}_{i1}^2=R(R-2x_{i}x-{2y}_{i}y+K_i)---\left(4\right)$

Wherein

deduct respectively r on (4) both sides ₁ ⁴(r ₁by formula (2), calculated), just can obtain

$d{p}_{i1}^2-{r_1}^4=R(K_i-2x_{i}x-{2y}_{i}y)---\left(5\right)$

Equation group (5) is write as to matrix form

Y＝AX (6)

Wherein

$Y=\left[\begin{array}{c}d{p}_{21}^2-{r_1}^4\\ \·\\ \·\\ \·\\ {\mathrm{dp}}_{N1}^2-{r_1}^4\end{array}\right],A=\left[\begin{array}{ccc}{K}_2& -x_2& -y_2\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ K_N& -x_N& -y_N\end{array}\right],X=\left[\begin{array}{c}R\\ 2\mathrm{xR}\\ 2\mathrm{yR}\end{array}\right].$

Further again, described step 5) in, (6) are solved by least-squares algorithm:

$\hat{X}={\left(A^{T}A\right)}^{-1}{A}^{T}Y---\left(7\right)$

The coordinate estimated value that obtains mobile station MS is:

${[x,y]}^T=\frac{{[\hat{X}\left(2\right),\hat{X}\left(3\right)]}^T}{2\hat{X}\left(1\right)}---\left(8\right)$

In formula (8)

representation vector

in k element, k=1,2,3.

Technical conceive of the present invention is: obtaining TOA between BS and MS (can be equivalent to distance) afterwards, just can pass through product of transformation, range finding is transformed into to the product of range finding, then build the positioning equation group according to the location geometrical relationship, after equation group is carried out to linearization process, can utilize the LS algorithm to be estimated the coordinate of MS.

TPOA refers to that the TOA by measurement is transformed into their product, the form multiplied each other in twos.Because the NLOS error has the trend risen along with the increase of actual distance, the performance of traditional location algorithm will be had a greatly reduced quality in this kind of environment, therefore must adopt certain method to eliminate the impact of NLOS error on positioning performance.The present invention utilizes this specific character of NLOS error, has proposed the wireless location algorithm based on TPOA, even in serious NLOS transmission environment, the present invention also can carry out the high-performance location to MS.In NLOS error Discussion of Linear Model (DDM, distance dependent model), the NLOS error rises along with the increase of actual distance, and the algorithm for estimating of this patent can effectively suppress the adverse effect of NLOS error for estimated performance.

Beneficial effect of the present invention is mainly manifested in: in the environment risen along with the increase of actual distance in the NLOS error, the present invention has realized the inhibition to the NLOS error under the prerequisite that does not increase computation complexity, with traditional LS class algorithm, compare, amount of calculation of the present invention does not increase, but is being significantly improved aspect the inhibitory action of NLOS error.

The accompanying drawing explanation:

The flow chart that Fig. 1 is the NLOS transmission environment wireless location method based on TPOA.

The schematic diagram that Fig. 2 is various algorithms positioning result in the DDM of NLOS mode.

Fig. 3 is in the DDM model, the schematic diagram of the maximum max of NLOS error on the impact of each arithmetic accuracy.

Embodiment

Below in conjunction with accompanying drawing, the invention will be further described.

With reference to Fig. 1～Fig. 3, a kind of NLOS transmission environment wireless location method based on TPOA comprises the following steps:

(1) receive by a plurality of base stations the signal that mobile station MS sends, and extract the TOA information in signal, and this information is pooled in dominant base;

(2) TOA measured is transformed into to the product TPOA of the time of advent;

(3) according to the location geometrical relationship, build the positioning equation group;

(4) the positioning equation group is carried out to mathematic(al) manipulation, be converted into system of linear equations;

(5) utilize least-squares algorithm to be estimated the coordinate of mobile station MS.

Described step 2) in, if utilize i base station BS _imeasure TOA _i, dominant base is made as the 1st base station, and the product of it and the 1st base station measurements is exactly TPOA so _i1, namely

TPOA _i1=TOA _i*TOA ₁ (1)；

Distance between mobile station MS and i base station (BS) is write as:

$r_i=\sqrt{{(x-x_i)}^2+{(y-y_i)}^2}={\mathrm{TOA}}_i\×c---\left(2\right)$

Wherein (x, y) and (x _i, y _i) be respectively the coordinate of mobile station MS and i BS, c is propagation velocity in the electromagnetic wave air;

According to above-mentioned formula, TPOA is transformed into to distance long-pending, if make dp _i1for the product between range finding, and without loss of generality, suppose BS ₁position be positioned at the origin of coordinates, so

${\mathrm{dp}}_{i1}^2=(x^2+y^2)+({(x-x_i)}^2+{(y-y_i)}^2)---\left(3\right).$

Further, described step 4) in, (3) are carried out to mathematic(al) manipulation

${\mathrm{dp}}_{i1}^2=R(R-2x_{i}x-2y_{i}y+K_i)---\left(4\right)$

Wherein

deduct respectively r on (4) both sides ₁ ⁴(r ₁by formula (2), calculated), just can obtain

${\mathrm{dp}}_{i1}^2-{r_1}^4=R(K_i-2x_{i}x-2y_{i}y)---\left(5\right)$

Equation group (5) is write as to rectangular

Y=AX (6)

Wherein

$Y=\left[\begin{array}{c}{\mathrm{dp}}_{21}^2-{r_1}^4\\ \·\\ \·\\ \·\\ {\mathrm{dp}}_{N1}^2-{r_1}^4\end{array}\right],A=\left[\begin{array}{ccc}{K}_2& -x_2& -y_2\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ K_N& -x_N& -y_N\end{array}\right],X=\left[\begin{array}{c}R\\ 2\mathrm{xR}\\ 2\mathrm{yR}\end{array}\right].$

Further again, described step 5) in, (6) are solved by least-squares algorithm:

$\hat{X}={\left(A^{T}A\right)}^{-1}{A}^{T}Y---\left(7\right)$

The coordinate that obtains mobile station MS is:

${[x,y]}^T=\frac{{[\hat{X}\left(2\right),\hat{X}\left(3\right)]}^T}{2\hat{X}\left(1\right)}---\left(8\right)$

In formula (8)

representation vector

in k element, k=1,2,3.

In Fig. 1, utilizing base station measurement at first to need to be converted into the form of measuring distance product after obtaining time of advent of signal, then set up the positioning equation group according to the mathematical relationship of this sum of products mobile station MS position, then by mathematic(al) manipulation, equation group is carried out to linearity and change processing, finally utilize the solution of equations that arrives of least-squares algorithm, and the coordinate of definite mobile station MS.

The emulation positioning result of the various algorithms of Fig. 2, black triangle represents base station location.This emulation adopts the topology of classical cellular network 7 base station models as locating base station, and wherein the coordinate of base station is respectively

in this emulation, get cellular cell radius r=1000m, and hypothesis range measurement error is the zero-mean gaussian variable that standard deviation equals 20m.Consider that in addition the NLOS error meets the DDM model, the NLOS error rises along with the increase of actual distance.The maximum that max is NLOS is decided to be 250m in this emulation.From 2 figure, can find out, positioning result of the present invention is more close to actual position.

Fig. 3 is root-mean-square error (RMSE, the root mean square error) comparison diagram of the present invention and other location algorithms, and its simulated environment is identical with Fig. 2, is mainly used in studying the impact of max on each arithmetic accuracy.The contrast algorithm comprise poor (TDOA, the time difference of arrival) algorithm time of advent, the time of advent and (TSOA, the time sum of arrival) algorithm that adopt the LS principle and the time of advent (TOA) algorithm.As can be seen from Figure 3, along with the increase of max, the precision of all algorithms is all descending, but different, and location algorithm decline of the present invention ground is the slowest, and its performance is better than other algorithm always.

Claims (4)

1. the NLOS transmission environment wireless location method based on TPOA, it is characterized in that: described localization method comprises the following steps:

1) receive by a plurality of base station BSs the signal that mobile station MS sends, and extract the TOA information in signal, and these information are pooled in the dominant base of location;

2) take the measured value of dominant base is reference, and the TOA of other base station measurements is multiplied by reference value, is transformed into TPOA;

3), according to the location geometrical principle, build and take the positioning equation group that TPOA is parameter;

4) the positioning equation group is carried out to linearization process;

5) utilize least-squares algorithm solving equation group, and position.

2. a kind of NLOS transmission environment wireless location method based on TPOA as claimed in claim 1, is characterized in that: described step 2), if utilize i base station BS _imeasure TOA _i, dominant base is made as the 1st base station, TOA _iwith the product of the 1st base station measurements be exactly TPOA _i1, namely

TPOA _i1＝TOA _i*TOA ₁ (1)；

Distance between mobile station MS and i base station BS is write as:

$r_i=\sqrt{{(x-x_i)}^2+{(y-y_i)}^2}={\mathrm{TOA}}_i\×c---\left(2\right)$

Wherein (x, y) and (x _i, y _i) be respectively the coordinate of mobile station MS and i BS, c is propagation velocity in the electromagnetic wave air;

According to above-mentioned formula, TPOA is transformed into to distance long-pending, if make dp _i1for the product between range finding, and without loss of generality, suppose BS ₁position be positioned at the origin of coordinates, so

${\mathrm{dp}}_{i1}^2=(x^2+y^2)({(x-x_i)}^2+{(y-y_i)}^2)---\left(3\right).$

3. a kind of NLOS transmission environment wireless location method based on TPOA as claimed in claim 2, is characterized in that: described step 4), (3) are carried out to mathematic(al) manipulation

${\mathrm{dp}}_{i1}^2=R(R-{2x}_{i}x-{2y}_{i}y+K_i)---\left(4\right)$

Wherein

r=x ²+ y ², on (4) both sides, deduct respectively r ₁ ⁴, r ₁calculated by formula (2), just can obtain

${\mathrm{dp}}_{i1}^2-{r_1}^4=R(K_i-2x_{i}x-{2y}_{i}y)---\left(5\right)$

Equation group (5) is write as to matrix form:

Y=AX (6)

Wherein

$Y=\left[\begin{array}{c}{\mathrm{dp}}_{21}^2-{r_1}^4\\ \·\\ \·\\ \·\\ {\mathrm{dp}}_{N1}^2-{r_1}^4\end{array}\right],A=\left[\begin{array}{ccc}{K}_2& -x_2& -y_2\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ K_N& -x_N& -y_N\end{array}\right],X=\left[\begin{array}{c}R\\ 2\mathrm{xR}\\ 2\mathrm{yR}\end{array}\right].$

4. a kind of NLOS transmission environment wireless location method based on TPOA as claimed in claim 3, it is characterized in that: described step 5), (6) are solved by least-squares algorithm:

$\hat{X}={\left(A^{T}A\right)}^{-1}{A}^{T}Y---\left(7\right)$

Then converting the coordinate estimated value that obtains mobile station MS is:

${[x,y]}^T=\frac{{[\hat{X}\left(2\right),\hat{X}\left(3\right)]}^T}{2\hat{X}\left(1\right)}---\left(8\right)$

In formula (8)

representation vector in k element, k=1,2,3.

Images