A method for improving positioning accuracy of UHF RFID tags based on SAR method

Abstract

The application discloses a method for improving positioning accuracy of UHF RFID tags based on a SAR method, and is applied to the field of RFID positioning and aims at the problem of poor positioning accuracy of UHF RFID tags based on the SAR method in the prior art. The method comprises the following steps: processing a measured phase of a tag backscattering signal, performing traditional tag position estimation based on the SAR, using continuous phase unwrapping, and solving a hyperbolic curve intersection point as a reference point. On the original measured phase, the reference point is used to realize fine phase unwrapping of a wider distance (several wavelengths) reader antenna position, and a larger focal length hyperbolic curve group is constructed on the basis. The joint phase unwrapping method of the application greatly improves the problem of multi-path error accumulation of traditional phase unwrapping. Therefore, a larger focal length hyperbolic curve can be constructed, the anti-noise capability of the positioning hyperbolic curve model itself is improved, and better positioning effect is obtained.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of RFID positioning, and particularly relates to a UHF RFID tag positioning technology based on a SAR method. BACKGROUND

[0002] With the development of the Internet of Things, RFID positioning technology has developed rapidly. Studies have shown that phase is more sensitive and stable to distance than received signal strength, and phase-based positioning technology has gradually become mainstream. Since there is a cycle number ambiguity in measuring phase, direct phase-based positioning methods are computationally complex and have low positioning performance. Therefore, phase unwrapping has emerged, and positioning technology using phase unwrapping methods has developed rapidly, and SAR is one of the important methods.

[0003] Prior art one: “P. Tripicchio, et al.,"A synthetic aperture UHF RFID localization method by phase unwrapping and hyperbolic intersection," IEEE T. Autom. Sci. Eng., vol. 19, no. 2, pp. 933-945, 2021.” proposes a positioning method that continuously measures to achieve phase unwrapping and uses discontinuous points to construct hyperbolas. By increasing the focal length of the hyperbola and using a cost function, the noise resistance of the hyperbolic model is improved, and good positioning results are obtained. However, in order to reduce the error of traditional phase unwrapping, a sampling interval of about 2 cm (much smaller than the theoretical sampling interval limit of 8.5 cm) is used to construct a hyperbola with a large focal length (minimum focal length of 15 cm), greatly increasing the workload during measurement. And for the positioning range of passive RFID (about 10 m), the focal length of the hyperbola is still too small, and its noise resistance performance still has room for improvement. This paper uses an 8 cm sampling interval to achieve a focal length of 50 cm with fewer sampling points, greatly improving the noise resistance performance of the model.

[0004] Prior art two: Patent application CN 118050681 A proposes that the carrying component makes driving motion according to a preset route indoors, and collects data information of all UHF RFID tags to be measured during the driving motion. Based on the data information and SAR positioning technology, the spatial coordinate model of the UHF RFID tag reading and collection is calculated and obtained.

[0005] Prior Art 3: Patent application CN 110850401 B proposes an RFID tag positioning method based on a motion model and synthetic aperture. By analyzing the motion model of the SAR scene, a convex optimization strategy is introduced into the target positioning process. The collected phase information is then phase-unwrapped and combined, and fixed frequency offsets are eliminated. Finally, the steepest descent method is used to optimize the similarity function. This method ultimately achieves high-precision positioning with extremely low computation time.

[0006] Prior Art 2 and 3 describe solutions that utilize antenna mobility for positioning. However, mobile antenna solutions are limited by specific scenarios, hindering widespread application and incurring additional costs. Implementing SAR positioning with fixed antenna arrays without compromising positioning accuracy could significantly improve the practicality of this technology, expand its application scenarios, and reduce its costs. Summary of the Invention

[0007] To solve the above technical problems, the present invention proposes a method for improving the positioning accuracy of UHF RFID tags based on the SAR method. On the original measurement phase, reference points are used to achieve fine phase unwrapping of the reader antenna position over a wider distance (several wavelengths), greatly improving the accumulation problem of multipath errors in traditional phase unwrapping.

[0008] The technical solution adopted by the present invention is: a method for improving the positioning accuracy of UHF RFID tags based on the SAR method, comprising:

[0009] S1, obtain the measured phase of the tag backscattered signal at different reader antenna positions;

[0010] S2. Calculate the measured phase difference values ​​collected from two reader antenna positions using the first distance, and calculate the difference in distance from the tag to each of the two reader antenna positions using the first distance based on the measured phase difference values;

[0011] S3. Using the absolute value of the distance difference obtained in step S2 as the real axis length, and using the two reader antenna positions using the first distance corresponding to the distance difference obtained in step S2 as foci to generate a hyperbola;

[0012] S4. Obtain the estimated position of the tag by finding the intersection of the hyperbola;

[0013] S5. Calculate the measured phase difference values ​​collected from the two reader antenna positions using the second distance, and calculate the difference in distance from the tag to each of the two reader antenna positions using the second distance based on the measured phase difference values; the second distance is greater than the first distance;

[0014] S6. Using the absolute value of the distance difference obtained in step S5 as the real axis length, and using the two reader antenna positions using the second distance corresponding to the distance difference obtained in step S5 as foci to generate a hyperbola;

[0015] S7. Calculate and obtain the final tag positioning result based on the hyperbola generated in step S6 that has the smallest distance from the estimated position obtained in step S4.

[0016] The beneficial effects of the present invention are as follows: it is required to carry the reader antenna by an electric-controlled guide rail or other mobile carrier, move along a predetermined linear or curved trajectory, and obtain the phase of the tag backscattered signal at different positions. Similarly, an antenna array can also be used to obtain the received signal phase through multiple antenna elements to simulate the effect of a mobile reader antenna. The method includes processing the measured phase of the tag backscattered signal, performing traditional SAR-based tag position estimation, using continuous phase unwrapping and solving the hyperbola intersection as a reference point. On the original measured phase, the reference point is used to achieve fine phase unwrapping of the reader antenna position at a wider distance (several wavelengths), and on this basis, the construction of a larger focal length hyperbola group is achieved.

[0017] This new joint phase unwrapping method significantly improves the accumulation of multipath errors in traditional phase unwrapping. This allows for the construction of a hyperbola with a larger focal length, improving the noise immunity of the positioning hyperbola model itself and achieving better positioning results. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a geometric diagram of the overall positioning solution of the present invention;

[0019] Figure 2 is the phase image of the present invention;

[0020] Among them, (a) is the measured phase data diagram, and (b) is the traditional phase unwrapping diagram;

[0021] Figure 3 This is a flow chart of the overall positioning algorithm of the present invention;

[0022] Figure 4 This is a schematic diagram of the positioning principle of the second part of the present invention;

[0023] Figure 5 This is a schematic diagram of the positioning principle of the third part of the present invention;

[0024] Figure 6 The comparison results between the method of the present invention and the prior art are shown;

[0025] Among them, (a) is the error cumulative distribution diagram, (b) is the box plot;

[0026] Figure 7 Outdoor positioning results of the method of the present invention;

[0027] Among them, (a) is the outdoor positioning result of the method of the present invention (26 positions) in actual measurement, and (b) is a box plot comparing the results of the method of the present invention (26 positions) and the method of the prior art (26 positions) in an outdoor environment in actual measurement;

[0028] Figure 8 The indoor positioning result of the method of the present invention;

[0029] Among them, (a) is the indoor positioning result of the method of the present invention (26 positions) in actual measurement, and (b) is a box plot comparing the results of the method of the present invention (26 positions) and the method of the prior art (26 positions) in an indoor environment in actual measurement;

[0030] Figure 9 Positioning results for the center array and two end antennas (16+2 positions);

[0031] Among them, (a) is the outdoor positioning result, (b) is the indoor positioning result;

[0032] Figure 10 Schematic diagram of the positioning scheme for the central array and the two end antennas (16+2 positions).

[0033] Explanation of reference numerals: 1 is the common plane of the tag and the reader antenna, 2 is the UHF RFID tag, 3 is the reader antenna, 4 is the phase of the received signal, 5 is the reader, 6 is the guide rail, 7 is the track, 8 is the interval Δ, 9 is the position of the reader antenna, 10 is the computer, 11 is the distance, 12 is the coaxial line, 13 is Figure 2 (b) The continuous phase, 14 is the tag position, 15 is the distance difference, 16 and 17 are two circles, 18 is all possible intersection points between the two circles, 19 is the hyperbola, 20 is the focal length, 21 is the center point of the hyperbola, 22 is a rough estimate of the tag position, 23 is multiple sets of hyperbolas, and 24 is A′ n =(d n -d N / 2+n ) / 2, 25 is 26 27 and 28 are abnormal values. DETAILED DESCRIPTION

[0034] To facilitate those skilled in the art to understand the technical content of the present invention, the present invention is further explained below with reference to the accompanying drawings.

[0035] The overall positioning method of the present invention is as follows Figure 1 Assume a two-dimensional UHF RFID positioning system, where positioning is performed on the common plane 1 of the tag and reader antenna, as shown in Figure 1 As shown (the system can be expanded to three-dimensional positioning by increasing the number of antenna elements in the vertical direction). Single or multiple UHF RFID tags 2 equipped with linearly polarized / circularly polarized antennas, and linearly polarized / circularly polarized reader antennas 3, which are connected to a commercial UHF RFID reader 5 with the function of measuring the phase of the received signal 4, are used to locate the tag 2. The reader antenna 3 is mounted on an electrically controlled guide rail 6 (the total length of the guide rail is L), which enables the reader antenna 3 to move in an orderly manner along any predefined trajectory 7 to meet the requirements of the Nyquist spatial sampling theorem (the spacing Δ8 is less than one-quarter wavelength of the frequency used for measurement). Any movable carrier, such as a robot or drone, can be used to sample the received signal phase measurement values ​​along a certain trajectory, provided that the Nyquist spatial sampling criterion is met. Alternatively, a fixed antenna array with elements can be used, with the unit spacing required to be less than one-quarter wavelength of the frequency used for measurement. According to UHF RFID standards and protocols, such as ISO18000-6C, the reader 5 performs an inventory at each position 9 of the reader antenna 3 to identify all possible tags 2 and collects measurement data of the phase 4 of the received signal at each reader antenna position 9 (for each tag separately), where the reader antenna position is S = {s1, s2, ... s n …s N}, s n =(s nx ,s ny ). Where N is the number of sampling points, which is determined by the sampling point interval and the length of the guide rail: The computer 10 controls the entire tag identification and measurement process of the tag phase 4, as well as the movement of the reader antenna 3 and continuous data acquisition. For each tag 2, the phase 4 measured by the reader is modeled as:

[0036]

[0037] where Φ n,i is the i-th measured phase at the n-th reader antenna position 9; λ is the wavelength (the UHF RFID global standard frequency band is 860MHz~960MHz, corresponding to a wavelength of 31.25cm~34.89cm); d n is the distance between the nth reader antenna position and the tag 11; is the initial phase, which is related to the reader antenna 3, the tag 2 antenna and the length of the coaxial line 12, and is usually considered to be a constant; is the phase error, which includes many factors, such as thermal noise and wireless channel multipath problems. As shown in formula (1), Figure 2As shown in (a), there is ambiguity in the number of whole cycles in the measured phase 4, which makes it impossible to directly obtain the distance 11 information from the measured phase information.

[0038] The method flow of the present invention is shown in Figure 3 , the first step uses the following phase unwrapping formula

[0039]

[0040] in is the floor function.

[0041] according to Figure 3 The algorithm shown in the first step is to Figure 2 The measurement phase 4 in (a) is expanded into Figure 2 (b) The continuous phase 13. This step is the same as the traditional UHF RFID SAR-based positioning technology. In actual testing, the model in formula (1) cannot fully describe the interference factors of phase measurement. In order to reduce the measurement error To evaluate the impact of abnormal data, it is necessary to collect multiple phase sample data Φ of each tag at each reader antenna position. n,i ,i∈{1,..,M} and use the phase average φ n =mean(Φ n,i ) and formula (2) for phase unwrapping, where M = 100, which is related to the actual environment and is used to analyze the statistical stability of the measurement data. Its value can be adjusted according to the actual environmental noise and multipath interference level. At the same time, calculate σ n =std(Φ n,i ), n∈{1,..,N}, that is, the standard deviation of the phase sample data at the sampling position (for Φ spanning the entire cycle n,i std(·) represents the standard deviation.

[0042] In the second step, the tag position 14 is roughly estimated using the expanded continuous phase value 13 and the method of solving the hyperbola intersection point. Figure 4 As shown, the phase difference between the discrete measurements is used The distance difference between the tag and the n,n+2th reader antenna position is 15, and the calculation formula is Here, we use discontinuously measured phase data. An increase in the interval will lead to an increase in phase unwrapping error (reduced reliability), but it can also increase the anti-noise effect of positioning (increased positioning accuracy). Since only a rough estimate is needed, the interval value in this invention is 2. For any reader antenna position pair, the tag position 14 is the intersection (x, y) of the two circles 16 and 17, and the difference 15 between the radii of the two circles is d n -d n+2, the two circles intersect at two points, one in the plane y>0 and the other in the plane y<0. Because the reader antenna is only in all possible intersection points 18 between the two circles in the plane y>0 (including the tag position (x, y); Figure 4 As shown. When a circle is drawn from two fixed points with a fixed radius difference, the trajectory formed by changing the radius is a hyperbola19, because a hyperbola is defined as the trajectory of points whose distance difference from two fixed points (called foci) is constant. By calculating the distance difference15 of all pairs of reader antenna positions, the calculation formula is We can get N-2 hyperbolas 19, and the labels are obviously located at their intersections 14, which are denoted as (x, y). However, due to the phase error The influence of , resulting in N-2 hyperbolas can not intersect at one point, need to use the method of minimizing the error to deal with. In addition, the hyperbola equation:

[0043]

[0044] where a n ,b n ,c n They represent the length of the real semi-axis 15 (d n+2 -d n =2a n ), length of the imaginary semi-axis, and focal length 20(c n =(s n+2 -s n ) / 2). Since the hyperbola equation is nonlinear, solving the intersection problem is too complicated. Therefore, the asymptote equations of the hyperbola are used to construct a linear system of equations (the error introduced is very small, less than the phase error, and can be ignored):

[0045]

[0046] in is the position of the center point of the hyperbola 21, is a constant whose sign depends on: Take the above symbols, Take the following symbols, for example when Take the above symbols, expressed as The error sources of phase measurement include thermal noise, multipath and other factors. It is reasonable to think that the position with smaller measurement standard deviation has higher confidence. Therefore, the sampling position weighting coefficient χ is introduced. n It is used to reduce the contribution of the sampling position that is significantly affected by the phase error, where the weighting coefficient of the asymptote is X n To construct the mean of the weighted coefficients of the two sampling positions of the hyperbola:

[0047]

[0048] The weighted linear equations are:

[0049]

[0050] Solve the linear equations to get the label position A rough estimate of .

[0051] Label Position The rough estimate of may be relatively poor because current methods cannot effectively handle interference caused by multipath and noise. It is clear from formula (2) that phase unwrapping leads to the accumulation of phase errors, resulting in large errors when calculating continuous phase unwrapping. Furthermore, in real environments, errors cannot be assumed to be uniformly distributed. There are issues such as multipath, the angle between the tag antenna and the reader antenna (in actual measurements, we found that the tag angle significantly affects the standard deviation of the phase measurement), and these issues can all lead to errors in phase unwrapping.

[0052] In fact, limiting the discontinuous phase measurement difference The main reason for the gap between the two locations is that continuous phase unwrapping produces large errors. A larger focal length can help reduce phase noise interference, thereby improving positioning accuracy. However, this approach is limited by the accumulation of phase unwrapping errors. To improve the accuracy of tag estimation, a third step is introduced, which involves combining phase unwrapping with the hyperbola intersection solution.

[0053] The third step is to start with the original average measured phase φ n (Not affected by the phase after phase expansion The error accumulation problem is affected by the use of non-continuous reader antenna positions with a larger distance, which is half the track length, φ n,N / 2+n =φ n -φ N / 2+n Due to the large distance between the reader antenna locations, the ambiguity in the number of phase cycles significantly affects the phase difference:

[0054]

[0055]

[0056]

[0057]

[0058] Among them, k n =-k n,max ,…,-2,-1,0,1,2,…k n,max , k n,maxThe distance difference d between the two positions of the hyperbola is constructed n -d N / 2+n Decide, satisfy:

[0059] (d n -d N / 2+n ) / (λ / 2)-1<k n,max <(d n -d N / 2+n ) / (λ / 2) (11)

[0060] in, Represents a set of integers.

[0061] However, we can now use the rough estimate of the label position from the second step 22 k pairs of periodic fuzzy numbers n Estimate. Assuming there is no error, the correct k n The constructed hyperbola should be estimated in the second step. Make an identity transformation on formula (10) and get d n -d N / 2+n =λ / 4π(φ n -φ N / 2+n -2πk n ), it can be seen that the distance difference d n -d N / 2+n Depends on k n The value of each pair of phase measurement values ​​φ n and φ N / 2+n Construct multiple sets of hyperbolas 23 (each set consists of a specific k n The value is determined, k′ n =-k n,max ,…,-2,-1,0,1,2,…k n,max ),like Figure 5 To determine each pair of phase measurements φ n and φ N / 2+n The correct k′ between n The value of the hyperbola is determined by roughly estimating the 22 and verify whether the coordinate belongs to or is close to one of these hyperbolas. From a mathematical point of view, this requires determining The minimum hyperbolic expression of the distance between different hyperbolic branches. As in (4) and (6), the minimum distance is calculated using asymptotes, which is mathematically expressed as:

[0062]

[0063] Among them, A′ n =(d n -d N / 2+n ) / 2 24, C′ n =(s n+N / 2 -s n ) / 2 25, by k n The value of is determined, where 26, the sign depends on: φ n -φ N / 2+n -2πk n ≥0 takes the above symbol, φ n -φ N / 2+n -2πk n <0 takes the following symbol. In determining the k corresponding to each hyperbola n and the related hyperbolic parameter A n ,B n ,C n Finally, the new set of hyperbolas (their asymptote equations) is used to construct a new weighted linear system of equations:

[0064]

[0065] This new combined phase unwrapping and hyperbola intersection solution significantly improves the noise and multipath error accumulation issues of traditional phase unwrapping. This allows for the construction of a hyperbola with a larger focal length, improving the noise immunity of the positioning hyperbola model and achieving better positioning results.

[0066] The present invention simulates and verifies the above positioning scheme: the simulation software is Mathworks Matlab, the simulation conditions are: L = 2m, Δ = 8cm (equivalent to N = 26 measurement positions), M = 100, frequency is 923MHz, normal noise, noise standard deviation is 5°, no multipath effect, and 1000 simulations are performed. The simulation results are as follows: Figure 6 As shown, Figure 6 (a) is the cumulative distribution diagram of positioning error, Figure 6 (b) is a box plot of positioning error. It can be seen that under the same simulation conditions, the method of the present invention is better than [P.Tripicchio, et al., "A synthetic aperture UHF RFID localization method byphase unwrapping and hyperbolic intersection," IEEE T.Autom.Sci.Eng., vol.19, no.2, pp.933-945, 2021] ( Figure 6-Figure 9 The algorithm in (represented as "existing technology") has greatly improved the positioning accuracy. Figure 6 (a), it can be seen that the error distribution of the method of the present invention is better, and the average positioning error is reduced by 60%. Figure 6(b) shows that the error distribution of the proposed method is more concentrated, with superior performance across all quantile metrics. The median error decreased by 70%, and the upper quartile error decreased by 57%. This is due to the proposed method's use of a joint phase unwrapping method, which expands the hyperbolic focal length without increasing the error of continuous phase unwrapping, thus enhancing the hyperbolic model's noise immunity.

[0067] At the same time, actual verification was carried out using a mobile antenna. In the actual test, an Impinj Speedway R420 reader, a commercial passive tag NXP U9, and a commercial guide rail with a length of L = 2m were used. The reader antenna was fixed and moved at intervals of Δ = 8cm (equivalent to N = 26 measurement positions) to achieve phase measurement in indoor and outdoor scenarios. Obviously, the number of target tag phase data measured at each position will affect the calculation of its phase mean and standard deviation. The present invention measured approximately M = 100 data at each measurement position. The frequency used in the test was 923MHz, and the wavelength was approximately λ = 32.5cm.

[0068] Figure 7 (a) and 8(a) show the positioning results of the algorithm of the present invention in outdoor and indoor environments, respectively (using all 26 locations); Figure 7 (b) and Figure 8 (b) Comparison of the positioning results of the proposed algorithm (using all 26 positions) and the most advanced UHFRFID tag SAR-based positioning algorithm [P.Tripicchio, et al., "A synthetic aperture UHF RFID localization method by phase unwrapping and hyperbolic intersection," IEEET.Autom.Sci.Eng., vol.19, no.2, pp.933-945, 2021]. Figure 7 In the comparison of outdoor positioning results in (b), the method of the present invention has no outliers (outliers) while the method of the paper has two outliers. The median positioning error decreased by 17%, and the upper quartile of the error decreased by 7%. Figure 8 (b) In the comparison of indoor positioning results, both methods have two outliers (27 and 28, due to severe multipath problems caused by proximity to a wall). In positioning at other locations, the proposed method significantly outperforms the paper's method. The median positioning error decreased by 33%, and the upper quartile error decreased by 50%. Furthermore, the indoor positioning results of the proposed method are very close to those of the outdoor positioning, indicating that the multipath problem is well addressed in this method.

[0069] Figure 9The positioning results of the algorithm of the present application (using 16 intermediate positions + 2 end positions, as shown in the schematic diagram Figure 10 ) and the algorithm in [P. Tripicchio, et al., "A synthetic aperture UHF RFID localization method by phase unwrapping and hyperbolic intersection," IEEE T. Autom. Sci. Eng., vol. 19, no. 2, pp. 933-945, 2021] (using all 26 positions) are compared. The method uses the middle array to determine the reference point, and uses the two end antennas to realize joint phase unwrapping, thereby improving the positioning accuracy. Figure 9 (a) In the outdoor positioning comparison, both methods have two abnormal values. In the case of reducing 8 measurement positions, the positioning results of the present application are better than the algorithm in [P. Tripicchio, et al., "A synthetic aperture UHF RFID localization method by phase unwrapping and hyperbolic intersection," IEEE T. Autom. Sci. Eng., vol. 19, no. 2, pp. 933-945, 2021]. The median of the error is reduced by 30%, and the upper quartile of the error is reduced by 26%. In Figure 9 (b) In the indoor positioning comparison, both methods have two abnormal values (27, 28), and in the case of reducing 8 measurement positions, the positioning results of the present application are slightly better than the algorithm in [P. Tripicchio, et al., "A synthetic aperture UHF RFID localization method by phase unwrapping and hyperbolic intersection," IEEE T. Autom. Sci. Eng., vol. 19, no. 2, pp. 933-945, 2021]. The median of the error is reduced by 4%, and the upper quartile of the error is reduced by 15%. It verifies the feasibility of the method in reducing the measurement units and realizing the fixed array SAR positioning method such as Figure 10 .

[0070] Those skilled in the art will appreciate that the embodiments described herein are intended to aid the reader in understanding the principles of the present invention, and it should be understood that the scope of the present invention is not limited to such specific descriptions and embodiments. Various modifications and variations are readily apparent to those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims.

Claims

1. A method for improving the positioning accuracy of UHF RFID tags based on SAR method, characterized in that: include: S1, obtain the measured phase of the tag backscattered signal at different reader antenna positions; S2. Calculate the measured phase difference values ​​collected from two reader antenna positions using the first distance, and calculate the difference in distance from the tag to each of the two reader antenna positions using the first distance based on the measured phase difference values; S3. Using the absolute value of the distance difference obtained in step S2 as the real axis length, and using the two reader antenna positions using the first distance corresponding to the distance difference obtained in step S2 as foci to generate a hyperbola; S4. Obtain the estimated position of the tag by finding the intersection of the hyperbola; S5. Calculate the measured phase difference value collected from the two reader antenna positions using the second distance, and calculate the periodic ambiguity number k using the estimated position obtained in step S4. n , according to the measured phase difference and k n Calculating the difference in distance from the tag to each of the two reader antenna positions using a second distance, wherein the second distance is greater than the first distance; Among them, A′ n is half of the second distance, C′ n is half the difference between the positions of the reader antennas at the two second distances, It is half of the sum of the positions of the reader antennas at the two second distances. The symbol "±" in the formula The determination method is: when the difference between each pair of phase measurement values ​​is greater than or equal to 2πk n , take the upper symbol; otherwise take the lower symbol; S6. Using the absolute value of the distance difference obtained in step S5 as the real axis length, and using the two reader antenna positions using the second distance corresponding to the distance difference obtained in step S5 as foci to generate a hyperbola; S7. Calculate and obtain the final tag positioning result based on the hyperbola generated in step S6 that has the smallest distance from the estimated position obtained in step S4.

2. The method for improving the positioning accuracy of UHF RFID tags based on the SAR method according to claim 1, characterized in that: Multiple phase sample data are collected for each tag at each reader antenna position.

3. The method for improving the positioning accuracy of UHF RFID tags based on the SAR method according to claim 2, characterized in that: The specific implementation process of step S4 is: The hyperbola equation generated in step S3 is recorded as: Among them, a n represents the length of the real semi-axis of the nth hyperbola, b n represents the length of the imaginary semi-axis of the nth hyperbola, c n represents the focal length of the nth hyperbola; Use the equations of the asymptotes of the hyperbola to construct a system of linear equations: Among them, a1 represents the length of the real semi-axis of the first hyperbola, b1 represents the length of the imaginary semi-axis of the first hyperbola, and c1 represents the focal length of the first hyperbola. represents the center point of the first hyperbola, a2 represents the length of the real semi-axis of the second hyperbola, b2 represents the length of the imaginary semi-axis of the second hyperbola, and c2 represents the focal length of the second hyperbola. represents the center point of the second hyperbola, a3 represents the real semi-axis length of the third hyperbola, b3 represents the imaginary semi-axis length of the third hyperbola, and c3 represents the focal length of the third hyperbola. Indicates the position of the center point of the third hyperbola, a n represents the length of the real semi-axis of the nth hyperbola, b n represents the length of the imaginary semi-axis of the nth hyperbola, c n represents the focal length of the nth hyperbola, Indicates the position of the center point of the nth hyperbola; the symbol "±" in the formula The determination method is: when the difference between the phase measurement values ​​of each pair of first distances is greater than or equal to 0, take the upper sign; otherwise, take the lower sign; Introducing the sampling position weighting coefficient χ n , according to χ n Get the weighting coefficient X of the asymptote n ; The weighted linear equations are: Solve the weighted linear equations to get the label position A rough estimate of .

4. The method for improving the positioning accuracy of UHF RFID tags based on the SAR method according to claim 3, characterized in that: Sampling position weighting coefficient χ n The calculation formula is: Among them, σ n Indicates the standard deviation of the phase sample data at the sampling position.

5. The method for improving the positioning accuracy of UHF RFID tags based on the SAR method according to claim 3 or 4, characterized in that: X n is the mean of the weighted coefficients of the two sampling positions to construct the hyperbola.

Images