Method for improving UHF RFID tag positioning precision based on SAR method

Abstract

The invention discloses a method for improving UHF RFID (Ultra High Frequency Radio Frequency Identification) tag positioning precision based on an SAR (Synthetic Aperture Radar) method, which is applied to the field of RFID positioning and aims to solve the problem of poor UHF RFID tag positioning precision based on the SAR method in the prior art. The method comprises the steps of processing a measurement phase of a label backscattering signal, performing traditional SAR-based label position estimation, using continuous phase unwrapping and solving a hyperbolic curve intersection point as a reference point. On the basis of an original measurement phase, fine phase unwrapping of a wider-distance (multiple wavelengths) reader-writer antenna position is realized by using a reference point, and construction of a larger-focal-length hyperbolic curve group is realized on the basis of the fine phase unwrapping; by adopting the combined phase unwrapping method provided by the invention, the problem of accumulation of multipath errors of traditional phase unwrapping is improved to a great extent. Therefore, the hyperbolic curve with a larger focal length can be constructed, the anti-noise capability of the positioning hyperbolic curve model is improved, and a better positioning effect is obtained.

Description

Technical Field

[0001] The invention belongs to the field of RFID positioning, and in particular relates to a UHF RFID tag positioning technology based on a SAR method. Background Art

[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 the received signal strength, and phase-based positioning technology has gradually become mainstream. Due to the ambiguity of the number of cycles in the measured phase, the positioning method directly based on the measured phase is complex to calculate and has poor positioning effect. Therefore, phase unwrapping has emerged, and positioning technology using the phase unwrapping method has developed rapidly, and SAR is one of the important methods.

[0003] Prior art 1: "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." proposes a positioning method that uses continuous measurement to achieve phase unwrapping and uses discontinuous points to construct a hyperbola. By increasing the focal length of the hyperbola and using the cost function, the anti-noise ability of the hyperbola model is improved, and a good positioning effect is obtained. However, in order to reduce the error of traditional phase unwrapping and construct a hyperbola with a large focal length (minimum focal length 15cm), a sampling interval of about 2cm (much smaller than the theoretical sampling interval limit of 8.5cm) is used in the paper, which greatly increases the workload during measurement. And for the passive RFID positioning range (about 10m), the focal length of the hyperbola is still too small, and its anti-noise performance still has room for improvement. This paper adopts an 8cm sampling interval and achieves a focal length of 50cm with fewer sampling points, which greatly improves the anti-noise performance of the model.

[0004] Prior art 2: The patent application with publication number CN 118050681 A proposes that a transport component moves indoors according to a preset route and collects data information of all UHF RFID tags to be tested during the movement. Based on the data information and SAR positioning technology, a spatial coordinate model of the reading collection of the UHF RFID tag is calculated and obtained.

[0005] Prior art 3: The patent application with announcement number CN 110850401 B proposes a RFID tag positioning method based on motion model and synthetic aperture. By analyzing the motion model of the SAR scene, the convex optimization strategy is introduced into the target positioning process, the collected phase information is phase-unwrapped, phase-combined, and fixed frequency offset is eliminated, and finally the steepest descent method is used to optimize the similarity function. Finally, high-precision positioning with extremely low computing time is achieved.

[0006] Prior art 2 and 3 describe solutions for achieving positioning by moving antennas, but the mobile antenna solution is limited by specific scenarios and is difficult to be widely used, and the mobile antenna requires additional costs. If fixed antenna array SAR positioning is achieved without reducing positioning accuracy, the practicality of the technology can be greatly improved, its application scenarios can be expanded, and its cost can be reduced. Summary of the invention

[0007] In order 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, the reference point is used to realize the fine phase unwrapping of the reader antenna position at a wider distance (several wavelengths), which greatly improves 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, obtaining the measured phase of the tag backscattered signal at different reader antenna positions;

[0010] S2, calculating the measured phase difference values ​​collected from two reader antenna positions using the first distance, and calculating the difference in distance from the tag to each of the two reader antenna positions using the first distance according to 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, obtaining the estimated position of the tag by finding the intersection of the hyperbola;

[0013] S5, calculating the measured phase difference values ​​collected from two reader antenna positions using the second distance, and calculating the difference in distance from the tag to each of the two reader antenna positions using the second distance according to the measured phase difference value; 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 the focus to generate a hyperbola;

[0015] S7. Calculate and obtain the final positioning result of the tag 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 through an electrically controlled rail or other mobile carrier, move along a predetermined linear or curved trajectory, and obtain the phase of the tag backscatter 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 backscatter 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 greatly improves the accumulation problem of multipath errors in traditional phase unwrapping. Therefore, a hyperbola with a larger focal length can be constructed, the anti-noise ability of the positioning hyperbola model itself can be improved, and better positioning results can be obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a geometrical schematic diagram of the overall positioning scheme 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 It is the overall positioning algorithm flow chart 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 of 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 result of outdoor positioning 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 result of indoor positioning 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] Fig. 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] Fig.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 outliers. 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). A single or multiple UHF RFID tags 2 equipped with linear / circular polarization antennas, a linear / circular polarization reader antenna 3, which is connected to a commercial UHF RFID reader 5 with a received signal phase 4 measurement function, is 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), and the guide rail 6 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 interval 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 value in a certain trajectory, but it is necessary to meet the Nyquist spatial sampling criterion. A fixed antenna array with elements can also be used, and the unit spacing needs 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 of the tag phase 4 process, as well as the movement of the reader antenna 3 and continuous data collection. For each tag 2, the phase 4 measured by the reader is modeled as follows:

[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 the wavelength: 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 an ambiguity in the number of integer 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 measured data. Its value can be adjusted according to the actual environmental noise and multipath interference level. At the same time, 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 unfolded 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 The distance difference between the tag and the antenna position of the nth, n+2th reader is 15, and the calculation formula is Here, the phase data of discontinuous measurement is used. The increase of the interval will lead to an increase in the 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 the present invention is 2. For any pair of reader antenna positions, 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 the radius difference between two fixed points is fixed, 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 difference of all reader antenna position pairs15, the calculation formula is We can get N-2 hyperbolas 19, and the labels are obviously located at their intersections 14, with the intersections represented by (x, y). However, due to the phase error The influence of , resulting in the N-2 hyperbolas cannot intersect at one point, need to be processed by minimizing the error. 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 ), the length of the imaginary semi-axis, and the focal length 20(c n =(s n+2 -s n ) / 2). Since the hyperbola equation is a nonlinear equation, it is too complicated to solve the intersection problem. Therefore, the asymptote equation of the hyperbola is used to construct a linear equation system (the error brought 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 Taking the above symbols, it is 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 sampling positions that are significantly affected by 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 location The rough estimation accuracy of may be relatively poor, because the current method cannot handle the interference caused by multipath and noise well. It can be clearly seen from formula (2) that phase unwrapping will lead to the accumulation of phase errors, resulting in large errors when calculating continuous phase unwrapping. In addition, the errors in the actual environment cannot be considered to be uniformly distributed. There are problems such as multipath, the angle between the tag antenna and the reader antenna (in actual measurements, we found that the tag angle will significantly affect the standard deviation of the phase measurement), which will cause errors in phase unwrapping.

[0052] In fact, limiting the discontinuous phase measurement difference The main reason for the interval between the two positions is that continuous phase unwrapping will produce large errors. A larger focal length may help reduce phase noise error interference, thereby improving positioning accuracy. However, this scheme is limited by the accumulation of phase unwrapping errors. In order to improve the accuracy of tag estimation, a third step is introduced, including joint phase unwrapping and 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 at 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 positions, 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 by 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 for periodic fuzzy numbers n Estimate. Assuming there is no error, the correct k n The constructed hyperbola should be estimated in the second step. By performing an identity transformation on formula (10), we can obtain 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 can be 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′ n value, and thus determine the correct hyperbolic branch, will use a rough estimate 22 and verify whether the coordinate belongs to or is close to one of these hyperbolas. Mathematically, 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 joint phase unwrapping and hyperbola intersection solution method greatly improves the noise error and multipath error accumulation problems of traditional phase unwrapping. Therefore, a hyperbola with a larger focal length can be constructed, improving the anti-noise ability of the positioning hyperbola model itself and obtaining 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 "prior art") 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) It can be seen that the error distribution of the method of the present invention is more concentrated, and it is better in all quantile indicators, among which the median error is reduced by 70% and the upper quartile error is reduced by 57%. This is because the method of the present invention uses a joint phase unwrapping method, which expands the hyperbolic focal length without increasing the continuous phase unwrapping error and enhances the anti-noise ability of the hyperbolic model.

[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 scenes. 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 algorithm of the present invention (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 due to severe multipath problems (27, 28, severe multipath problems caused by being close to the wall). In the positioning of other locations, the method of the present invention is much better than the method of the paper. The median positioning error decreased by 33%, and the upper quartile of the error decreased by 50%. In addition, the indoor positioning results of the method of the present invention are very close to the outdoor positioning results, indicating that the multipath problem is well handled in this method.

[0069] Fig. 9The algorithm of the present invention (using 16 intermediate positions + 2 end positions, the schematic diagrams of which are shown in FIG. Fig.10 The positioning results of the algorithm in [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] (using all 26 positions) are compared. This method uses the middle array to determine the reference point and uses the antennas at both ends to achieve joint phase unwrapping to improve positioning accuracy. Fig. 9 (a) In the outdoor positioning comparison, both methods have two outliers. When the number of measurement locations is reduced by 8, the positioning result of the proposed solution is better than the algorithm in [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]. Its median error decreased by 30%, and the upper quartile error decreased by 26%. Fig. 9 (b) In the indoor positioning comparison, both methods have two outliers (27, 28). When the number of measurement positions is reduced by 8, the positioning result of the proposed solution is slightly better than that of the algorithm in [P.Tripicchio, et al., "A synthetic aperture UHFRFID localization method by phase unwrapping and hyperbolic intersection," IEEE T.Autom.Sci.Eng., vol.19, no.2, pp.933-945, 2021]. The median error decreased by 4% and the upper quartile error decreased by 15%. This verifies that this method can reduce the number of measurement units and implement fixed array SAR positioning methods such as Fig.10 The feasibility is shown.

[0070] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. For those skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for improving the positioning accuracy of UHF RFID tags based on SAR method, characterized in that: include: S1, obtaining the measured phase of the tag backscattered signal at different reader antenna positions; S2, calculating the measured phase difference values ​​collected from two reader antenna positions using the first distance, and calculating the difference in distance from the tag to each of the two reader antenna positions using the first distance according to 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, obtaining the estimated position of the tag by finding the intersection of the hyperbola; S5, calculating the measured phase difference values ​​collected from two reader antenna positions using the second distance, and calculating the difference in distance from the tag to each of the two reader antenna positions using the second distance according to the measured phase difference values; The second distance is greater than the first distance; 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 the focus to generate a hyperbola; S7. Calculate and obtain the final positioning result of the tag 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 asymptote equations of the hyperbola to construct a linear system of 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 real semi-axis length of the second hyperbola, b2 represents the imaginary semi-axis length 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, 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 obtain 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. A method for improving the positioning accuracy of UHF RFID tags based on SAR method according to claim 3 or 4, characterized in that: X n It is the mean of the weighted coefficients of the two sampling positions to construct the hyperbola.

6. The method for improving the positioning accuracy of UHF RFID tags based on the SAR method according to claim 5, characterized in that: In step S7, the asymptote is used to calculate the minimum distance between the estimated position of the tag and the hyperbola generated in step S6: Among them, A′ n is half of the second distance, C′ n is half of 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 measurements is greater than or equal to 2πk n , take the upper sign; otherwise take the lower sign.

Images