CN112055306B - Reference base station selection method for reducing TDOA (time difference of arrival) positioning error - Google Patents

Abstract

The invention belongs to the technical field of indoor positioning, and particularly relates to a reference base station selection method for reducing TDOA positioning errors. The method solves the problem of how to select the reference base station in the mixed environment of LOS and NLOS, and avoids the extreme condition that all TDOA values contain NLOS errors due to the errors of the reference base station NLOS, thereby ensuring the measurement accuracy of the reference base station and further improving the positioning accuracy. The reference base station selection method for reducing the TDOA positioning error can identify the base station with smaller measurement error, and the TDOA positioning error can be reduced by taking the base station as the reference base station.

Description

Reference base station selection method for reducing TDOA (time difference of arrival) positioning error

Technical Field

The invention belongs to the technical field of indoor positioning, and particularly relates to a reference base station selection method for reducing TDOA positioning errors.

Background

With the advancement of communication technology and positioning technology, location-based services gradually penetrate into people's productive lives and become an indispensable part. Since more than 80% of human activities are performed indoors, indoor positioning technology has been the focus of research. The positioning algorithms are many in variety, most are based on ranging, and the ranging-based positioning algorithms include TOA (time of Arrival), TDOA (Angle of Arrival), and rssi (received Signal Strength index), wherein TOA and TDOA are the most commonly used positioning methods due to their simplicity, accuracy, and reliability.

TDOA location is performed by measuring the time difference of arrival of signals from a mobile station to different base stations to determine the difference in the distance of the mobile station to the different base stations. One of the methods for acquiring the TDOA signal is to ensure time synchronization between the base station and the mobile station, first acquire the time of arrival (TOA) of the signal from the mobile station to a single base station, and obtain the TDOA by subtracting the TOAs of other base stations from the reference base station with a certain base station as a reference base station and the TOA of the reference base station as a reference base station. Therefore, the measurement error of the reference base station directly affects the error of all the TDOA values, and the measurement accuracy of the reference base station has much larger influence on the positioning result than other base stations. In a complex indoor environment, NLOS (None-line-of-sight) condition is easy to occur, since the measurement error of the shielding base station of a building may increase by several orders of magnitude, if the NLOS condition occurs at the reference base station, this will cause all TDOA values to contain NLOS error, which is much larger than the positioning error when NLOS occurs at other base stations. The best choice of reference base station is the smallest measurement error among all base stations. In a complex indoor environment, the situation that the measurement error of one base station is always minimum and NLOS never occurs can not be guaranteed, but the base station with the smaller measurement error can be found out as a reference base station firstly during each positioning. Therefore, under the mixed environment of LOS and NLOS, how to select the reference base station of TDOA location has important significance.

Current research is based on how to identify anomalous measurements that contain NLOS errors. In terms of NLOS identification, most of identification can only identify one base station under the NLOS, the identification rate is not high enough in the case of multiple NLOS, and if the identification is wrong or all NLOS base stations cannot be identified, the reference base station still has the possibility of NLOS. NLOS recognition often requires setting a threshold, different environmental thresholds often differ, and how to select the threshold is also a problem. Pak et al use distributed and ellipsoid gating techniques to identify NLOS, threshold values are selected from chi-square table, (Jung Min Pak, Choon Ki Ahn, Peng Shi. distributed Hybrid/FIR Filtering for mixing NLOS Effects in TOA-Based Localization Using Wireless Sensor Networks [ J ] IEEE Transactions on Industrial Electronics, vol.64, No.6, June 2017.) this method can only deal with the situation that one base station is in NLOS, when a plurality of base stations are in NLOS, the method can not correctly identify the measured value containing NLOS error. Lee et al have studied a Kalman filtering Based NLOS identification method (Kyunghyun Lee, Jungkeun Oh, Kwanho you. closed-Form Solution of TDOA-Based coordination and Tracking: A secure Weighted Least Square Approach [ J ] Wireless Pers Commun (2017)94: 3451-3464), which can identify a plurality of NLOS, but has an assumption that there must be a known LOS base station. In summary, in the environment of mixing LOS and NLOS, there is no good method to guarantee the measurement accuracy of the reference base station at present.

Disclosure of Invention

The invention aims to solve the problem of how to select a reference base station under the mixed environment of LOS and NLOS, avoid the extreme condition that all TDOA values contain NLOS errors due to the errors of the reference base station NLOS and provide a reference base station selection method for reducing the TDOA positioning errors.

The purpose of the invention is realized by the following technical scheme: n base stations are arranged in the positioning area, N is more than or equal to 6, and the coordinates of each base station are known as BS _i ＝(x _i ,y _i ) 1,2, ·, N; the method comprises the following steps:

step 1: obtaining base stations BS _i Is measured by a set of measured values P ═ { d ═ d ₁ ,d ₂ ,...,d _N }; the base station BS _i Measured value d of _i By base station BS _i The measured TOA is multiplied by the propagation speed of the signal to obtain the signal;

step 2: obtain a subset S of the set P with 3 elements ₁ ,S ₂ ,...,S _M }，

And step 3: using maximum likelihood algorithm, from each set S _k The 3 measured values and corresponding base station coordinates contained in the same estimate the position of the mobile station

k＝1,2,...,M；

And 4, step 4: calculating each base station BS _i Measured value d of _i Accumulated residual SRes of _i ；

And 5: all measured values d _i According to the cumulative residual SRes _i Sorting the sizes in ascending order and putting the sorted sizes into an array Q;

step 6: select array Q middle front

Constructing a set T by using the measurement values; repeating the steps 2 to 5 once when P is equal to P and N is equal to T;

and 7: and selecting the base station corresponding to the first measured value in the array Q as the reference base station.

The invention has the beneficial effects that:

the invention provides a reference base station selection method for reducing TDOA positioning error, which can identify a base station with smaller measurement error, and can reduce the TDOA positioning error by taking the base station as a reference base station. The method solves the problem of how to select the reference base station in the mixed environment of LOS and NLOS, and avoids the extreme condition that all TDOA values contain NLOS errors due to the errors of the reference base station NLOS, thereby ensuring the measurement accuracy of the reference base station and further improving the positioning accuracy.

Drawings

FIG. 1 is a flow chart of a method of reference base station selection for reducing TDOA positioning error.

Detailed Description

The invention is further described below with reference to the accompanying drawings.

Under the mixed environment of LOS and NLOS, no good method is available at present to ensure the measurement accuracy of the reference base station, the invention provides a reference base station selection method for reducing the TDOA positioning error, the method can identify the base station with smaller measurement error, and the TDOA positioning error can be reduced by taking the base station as the reference base station. The invention aims to solve the problem of how to select the reference base station in the mixed environment of LOS and NLOS, and avoid the extreme condition that all TDOA values contain NLOS errors due to the errors of the reference base station NLOS, thereby ensuring the measurement accuracy of the reference base station and further improving the positioning accuracy.

The main idea of the invention is to evaluate the measurement error of each base station by accumulating residual errors, and select the base station with smaller measurement error as a reference base station. Residual refers in mathematical statistics to the difference between the actual observed value and the estimated value. The cumulative residual refers to the cumulative sum of residuals calculated a plurality of times.

The technical scheme adopted by the invention is as follows:

n base stations are arranged in the positioning area, and N is more than or equal to 6. The location coordinates of the base station are known and denoted as BS _i ＝(x _i ,y _i ) N. the mobile station position coordinates to be located are unknown, denoted as MS (x, y). One TDOA signal is associated with two base stations, and it is difficult to determine whether the ranging of one base station is accurate or not through TDOA, so the present invention uses TOA to determine which base station has smaller measurement error, and uses this base station as the reference base station to perform TDOA positioning. Base station BS _i The measured TOA times the propagation velocity of the signal gives a measure d of the distance of the mobile station from the base station _i 。

The selection process of the reference base station for reducing the TDOA positioning error is as follows:

1) a subset of the measurements of the base station is evaluated. Set P ═ { d ═ d ₁ ,d ₂ ,...,d _N All subsets of 3 element numbers are S ₁ ＝{d ₁ ,d ₂ ,d ₃ },S ₂ ＝{d ₁ ,d ₂ ,d ₄ },…,S _M ＝{d _N-2 ,d _N-1 ,d _N Therein of

2) The subset is used to calculate the mobile station position. Using maximum likelihood algorithm, for arbitrary sets S _k Wherein k ∈ [1, M ]]Using sets S _k And corresponding base station coordinates estimate the location of the mobile station

3) And solving the residual error of the measured value. For arbitrary sets S _k And an arbitrary measured value d _i Where k is [1, M ]]，i∈[1,N]If d is _i ∈S _k Measured value d _i Residual Res of _i (k) The calculation is as follows:

if it is not

Res _i (k)＝0。

4) The cumulative residual of the measurements is calculated. Calculating the measured value d _i Accumulated residual SRes of _i Where i ∈ [1, N ]]。

5) The measurements are sorted in ascending order of accumulated residual. Newly building an array Q, and taking all measured values d _i Put into Q, according to the cumulative residual SRes _i In ascending order of magnitude, where i e [1, N]。

6) The above steps are repeated once using the smaller half of the accumulated residuals in the measurements. Selecting a front in array Q

A new set T is constructed for each measurement, and the steps are repeated once.

7) And selecting the base station corresponding to the measurement value with the minimum accumulated residual as the reference base station. The base station corresponding to the first measurement value in the array Q is selected as the reference base station.

Example 1:

the invention considers the influence of signal non-line-of-sight propagation in indoor environment, identifies the base station with smaller measurement error by accumulating residual errors, avoids the reference base station measurement value containing NLOS error, and reduces the error of TDOA positioning. The invention provides a reference base station selection method for reducing TDOA positioning error, which is characterized in that a base station with smaller measurement error is identified as a TDOA reference base station in a mixed environment of LOS and NLOS, and the positioning error is reduced.

The invention provides a selection method of a reference base station for reducing TDOA (time difference of arrival) positioning errors, which takes two-dimensional positioning of 8 base stations as an example, wherein the base stations are respectively BS (base station) ₁ ,BS ₂ ,...,BS ₈ The coordinates of the base station are known, and the distances from the MS to the base station are d ₁ ,d ₂ ,...,d ₈ 。

1) Obtaining a set P ═ d ₁ ,d ₂ ,...,d ₈ All subsets of 3 elements, S ₁ ＝{d ₁ ,d ₂ ,d ₃ }，S ₂ ＝{d ₁ ,d ₂ ,d ₄ }，…，S ₅₆ ＝{d ₆ ,d ₇ ,d ₈ }，

2) Using maximum likelihood algorithm, for arbitrary sets S _k Where k ∈ [1,56 ]]Using sets S _k And corresponding base station coordinates estimate the location of the mobile station

3) For arbitrary sets S _k And a measured value d _i Wherein k ∈ [1,56 ]]，i∈[1,8]；

If d is _i ∈S _k Measured value d _i Residual Res of _i (k) The calculation is as follows:

if it is not

Res _i (k)＝0。

4) Calculating the measured value d _i Accumulated residual SRes of _i Where i ∈ [1,8 ]]，

5) Newly building an array Q, putting all measured values in the set P into Q, and accumulating residual errors SRes _i In ascending order of magnitude, where i e [1,8 ]]。

6) The first 4 measurements in the array Q are selected to construct a new set T, P ═ P ∞ T, and the above steps are repeated once.

7) The base station corresponding to the first measurement value in the array Q is selected as the reference base station.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (1)

1. A reference base station selection method for reducing TDOA positioning error is characterized in that N base stations are arranged in a positioning area, N is more than or equal to 6, and the coordinates of each base station are known as BS _i ＝(x _i ,y _i ) 1,2, ·, N; the method is characterized by comprising the following steps:

step 1: obtaining base stations BS _i Is measured by a set of measured values P ═ { d ═ d ₁ ,d ₂ ,...,d _N }; the base station BS _i Measured value d of _i By base station BS _i The measured TOA is multiplied by the propagation speed of the signal to obtain the signal;

step 2: obtain a subset S of the set P with 3 elements ₁ ,S ₂ ,...,S _M }，

And 3, step 3: using maximum likelihood algorithm, from each set S _k The 3 measured values and corresponding base station coordinates contained in the same estimate the position of the mobile station

And 4, step 4: calculating each base station BS _i Measured value d of _i Accumulated residual SRes of _i ；

And 5: all measured values d are compared _i According to the cumulative residual SRes _i Sorting the sizes in ascending order and putting the sorted sizes into an array Q;

step 6: select array Q middle front

Constructing a set T by using the measurement values; repeating the steps 2 to 5 once when P is equal to P and N is equal to T;

and 7: the base station corresponding to the first measurement value in the array Q is selected as the reference base station.

Images