RFID tag positioning system and method based on near-field antenna and tag positioning algorithm

Abstract

The present invention discloses an RFID tag positioning system and method based on a near-field antenna and tag positioning algorithm. The system primarily comprises an RFID tag attached to an item to be managed, an RFID reader / writer for reading or writing the RFID tag, an antenna array module connected to the RFID reader / writer, a data processing module, and a human-computer interaction module. RFID tags are used to identify items. The near-field antenna array module, through a dual-π-shaped feed point design, achieves uniform distribution of field strength within a specific cabinet area. A three-dimensional spatial model is determined based on the cabinet area and antenna distribution, forming a tag data information table. Combined with a precise positioning algorithm, the system achieves precise positioning of multiple tags in the described application scenario, effectively resolving the problems of missed reads and cross-reading during RFID tag inventory, and significantly improving tag positioning accuracy.

Description

Technical Field

[0001] The present invention belongs to the technical field of RFID, and in particular to an RFID tag positioning system and method based on a near-field antenna and a tag positioning algorithm. Background Art

[0002] With the widespread adoption of RFID tags, RFID technology is being used in an increasing number of scenarios, such as dynamic warehouse material inflow and outflow management and static document storage management. During use, an RFID tag is typically affixed to the object being managed. An RFID reader and antenna emit electromagnetic waves that cover the area where the RFID-tagged object is located, enabling communication between the reader and the tag and completing the tag reading. However, in actual use, the electromagnetic waves emitted by the RFID reader and antenna may reach locations other than the tag being read, making cross-reading and missed readings a common problem.

[0003] Specifically, (1) cross-reading refers to the area covered by the electromagnetic waves emitted by the antenna, which enables the RFID reader to read not only the tag of the object to be read, but also the tags at other locations around it. The reason for cross-reading is: the field strength distribution characteristics of the antenna. Different antennas have different field strength distribution ranges and strengths. As a result, the boundaries of the field strength area cannot be completely fixed. Therefore, when used in actual environments, it is inevitable that the reader will read surrounding tags, causing cross-reading. When other factors remain unchanged, when the antenna power is higher, the possibility of cross-reading is greater, and when the antenna power is lower, the possibility of cross-reading is lower.

[0004] (2) Missed reads: Due to the physical characteristics of RFID technology, when multiple tags completely overlap or the overlap rate is very high, the reader cannot read / count all target object tags. The higher the antenna power, the lower the missed read rate, and vice versa.

[0005] Therefore, cross-reading and missed reads are inherently conflicting factors. Increasing antenna power reduces missed reads, but cross-reading becomes more pronounced. Reducing antenna power improves cross-reading, but miss reads increase. Therefore, it's necessary to fundamentally improve the antenna's field strength distribution, avoiding weak and blind spots, ensuring a uniform field strength distribution and a controllable range.

[0006] Especially in the storage and positioning management of items, if the label of each item can be accurately read and the location of each item can be accurately located, the management efficiency of these items can be greatly improved.

[0007] In the prior art, such as the Chinese patent publication number CN111079874A, "A Modular Tag Positioning System and Method Based on RFID Technology," the disclosed modular tag positioning method, in which the RFID reader uses the antenna array module to count tags. Often, one tag is read by multiple antennas, resulting in interference values. When calculating the tag position, it is necessary to combine the number of times the tag is counted with the antenna's count result data to filter and determine the tag's location. However, the screening process of this method is as follows: if the number of times a certain antenna has counted the tag is less than a set threshold, the count result of the antenna is discarded. Setting the threshold in this operation is a very critical operation. If the threshold is set improperly, data that should not be discarded will be discarded, or data that should be discarded will be included in the data table, resulting in excessive inaccurate data being mixed in during the subsequent parabola generation process, which in turn leads to inaccurate final tag positioning. Summary of the Invention

[0008] In order to overcome the shortcomings of the existing technology, the present invention provides an RFID tag positioning system based on a near-field antenna and a tag positioning algorithm. The RFID tag positioning system can achieve accurate positioning of tagged items and meet the positioning needs in similar scenarios in different industries.

[0009] To achieve the above objectives, the present invention adopts a technical solution: an RFID tag positioning system based on a near-field antenna and a tag positioning algorithm, comprising an RFID tag set on an item to be managed, an RFID reader / writer for reading or writing the RFID tag, an antenna array module connected to the RFID reader / writer, a data processing module, and a human-computer interaction module, characterized in that the data processing module includes a tag positioning algorithm module, an antenna array three-dimensional space generation module, and a tag data information table generation module.

[0010] Arrange the antenna array module in the designated positioning area, partition the designated positioning area into multiple spatial positions, and number these spatial positions in sequence, called spatial points. Each antenna corresponds to a spatial point. Connect each antenna array to the antenna splitter of the RFID reader, connect the antenna array module to the RFID reader, and connect the RFID reader to the host computer running the human-computer interaction module and the data processing module.

[0011] The user operates the system through the human-computer interaction module to locate the items with RFID tags placed in the positioning area. After the RFID reader receives the command from the host computer to locate the item, it traverses each antenna on each antenna array to count the tags, obtains the RSSI signal value of each tag in the positioning area when it is counted, the number of times each tag is counted by each antenna per unit time, and the antenna number of the tag, and inputs it into the data processing module running on the host computer.

[0012] The antenna array three-dimensional space generation module in the data processing module forms a three-dimensional space model of the antenna array according to the spatial point position of each antenna in the antenna array, wherein the antennas are distributed in the spatial point position based on the antenna feed point, and the closer the spatial point position of the antenna is to the center of the three-dimensional space model, the better its position.

[0013] The tag data information table generation module in the data processing module generates a tag data information table for each tag based on the RSSI signal value each time the tag is counted, the number of times each tag is counted by each antenna per unit time, and the antenna number of the tag. The RSSI signal values ​​in the tag data information table are sorted and calculated according to their strength to obtain an RSSI sorting table of the tag data information table.

[0014] The tag positioning algorithm module in the data processing module integrates the RSSI sorting table of the tag data information table and the advantages and disadvantages of the spatial point position of the tag's antenna number in the three-dimensional spatial model of the antenna array, and then outputs the optimal antenna spatial point position as the positioning position of the tag and transmits it to the host computer.

[0015] Furthermore, the antenna array module is composed of a plurality of traveling wave antennas based on a π-type structure, and the traveling wave antenna includes a dual feed point structure.

[0016] Furthermore, the traveling wave antenna includes a PCB board, on which two groups of π-shaped structure traces are distributed, and each end of the two groups of π-shaped structure traces is independently provided with a feeding point, and the two feeding points are distributed on the left and right, and the two groups of π-shaped structure waveforms complement each other in spatial distribution.

[0017] Another invention purpose: The present invention also provides an RFID tag positioning method based on an RFID tag positioning system, comprising the following steps: ① Arrange antenna array modules in a designated positioning area, partition the designated positioning area into multiple spatial positions, and number these spatial positions in sequence, which are called spatial points. Each antenna corresponds to a spatial point, and each antenna array is connected to the antenna splitter of the RFID reader; ② After the RFID reader receives a command from the host computer to locate and inventory items with RFID tags in the positioning area, it completes the data inventory of the tags in the positioning area, obtains the RSSI signal value of the tag in the positioning area each time it is inventoried, the number of times the tag is inventoried by each antenna per unit time, and the antenna number of each tag inventoried, and inputs them into the data processing module running on the host computer; It is characterized in that ③ The spatial point position of each antenna in the antenna array is input into the antenna splitter of the data processing module A three-dimensional spatial model of the antenna array is formed in the linear array three-dimensional space generation module, in which the antennas are distributed in spatial points based on the antenna feed point, and the closer the spatial point of the antenna is to the center of the three-dimensional spatial model, the better its position; ④ The RSSI signal value of each tag each time it is counted, the number of times each tag is counted by each antenna per unit time, and the antenna number of the tag counted are input into the tag data information table generation module in the data processing module to generate a tag data information table for each tag, and after sorting and calculating according to the strength of the RSSI signal value in the tag data information table, an RSSI sorting table of the tag data information table is obtained; ⑤ The tag positioning algorithm module in the data processing module performs tag positioning processing based on the advantages and disadvantages of the spatial point of the antenna in the three-dimensional spatial model of the antenna array and the RSSI sorting table of the tag data information table, and then outputs the optimal antenna spatial point as the positioning position of the tag and transmits it to the host computer.

[0018] Furthermore, in step ⑤, the tag positioning algorithm processing includes the following processes: (1) If the RSSI signal value of the tag read by a certain antenna number and the ranking of the number of times the tag is read are both optimal in the RSSI sorting table of the tag data information table, then the spatial point where the antenna number in the tag data information table is located is output as the positioning position of the tag; (2) If the RSSI signal value of the tag read by a certain antenna number and the ranking of the number of times the tag is read are not both optimal in the RSSI sorting table of the tag data information table, then the spatial point where the antenna number in the optimal tag data information table is located is output as the positioning position of the tag according to the weighted total score ranking of the RSSI signal value, the number of times the tag is read and the spatial point of the antenna number in the tag data information table.

[0019] Furthermore, when the situation (2) occurs in step ⑤, the two antenna numbers corresponding to the first two RSSI signal values ​​are selected from the RSSI sorting table of the tag data information table, and if the spatial point positions of the two antenna numbers are consistent, the spatial point position of the antenna is output as the positioning position of the tag; if the spatial point positions of the two antenna numbers corresponding to the first two RSSI signal values ​​in the RSSI sorting table of the tag data information table are inconsistent, when one of the RSSI signal values ​​is greater than the other RSSI signal value by a set threshold value ω, the spatial point position of the antenna number corresponding to the RSSI signal value with a larger value is selected as the positioning position of the tag; If the difference between the first two RSSI signal values ​​in the RSSI sorting table of the tag data information table is less than the set threshold ω, the two read times corresponding to the two RSSI signal values ​​are selected for comparison. When the read times of one of them is higher than the other by a set threshold σ, the spatial point position of the antenna number corresponding to the larger read times is selected as the positioning position of the tag; if the difference between the two read times is less than the set threshold σ, the spatial point position of the optimal antenna number is output as the positioning position of the tag according to the total weight score of the RSSI signal values, read times and spatial points of the antenna numbers in the two tag data information tables.

[0020] Furthermore, the weighted total score in step ⑤ is calculated according to the weight formula: weight (antenna number spatial point) × coefficient (antenna number spatial point) + weight (RSSI) × coefficient (RSSI) + weight (number of times read) × coefficient (number of times read) = total score, wherein weight (antenna number spatial point) > weight (RSSI) > weight (number of times read), and the two tag data information tables are sorted by the total scores calculated according to the weight formula, and the spatial point where the antenna number in the tag data information table with the highest total score is located is selected as the tag positioning position.

[0021] Furthermore, the antenna array adopts a traveling wave antenna based on a π-type structure, which includes a PCB board, on which two groups of π-type structure traces are distributed, and a feeding point is independently provided at one end of each group of π-type structure traces. The two feeding points are distributed on the left and the right, and the two groups of π-type structure waveforms complement each other in spatial distribution.

[0022] Furthermore, the PCB board includes a PCB front panel and a PCB ground panel. The PCB front panel is distributed with two antenna copper-clad parts with corresponding π-shaped structure routing. A feeding point is provided at one end of each of the two antenna copper-clad parts, and the two feeding points are distributed on the left and right sides.

[0023] Using the above scheme, the PCB near-field antenna array employed in the present invention is composed of multiple (or more than two) traveling-wave antennas. Traveling-wave antennas have low radiation efficiency, effectively controlling the antenna's radiated energy within a short range, i.e., the antenna's near-field region. The PCB antennas are arranged in a dual-antenna configuration on a single PCB, with dual feed points extending from both sides in a π-shaped structure. The dual π-shaped structures complement each other, ensuring uniform distribution of antenna radiated energy in the near-field region. This fully controllable energy coverage effectively avoids the blind spots and weak areas associated with single-feed antennas. The antenna array is spread flat in the positioning area, so that the antenna field strength in the area is evenly distributed. Based on the antenna array, the tag signal strength read by the passive RFID reader is evenly distributed, avoiding missed readings and obvious cross-reading between antenna arrays in adjacent areas. In addition, for the interference value caused by a single tag being read by multiple antennas, the present invention determines a three-dimensional spatial model according to the actual shape of the positioning area (specific cabinet) and the antenna distribution, forms a tag data information table, and combines it with a precise positioning algorithm to achieve precise positioning of multiple tags in the application scenario, effectively solving the problems of missed reading and cross-reading in the RFID tag inventory process, and greatly improving the positioning accuracy of the tags.

[0024] The present invention will be further described below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Attachment Figure 1 This is a schematic diagram of the dual-feed point near-field antenna structure of the present invention, where (numbered, the same below) 1 is the PCB front panel of the near-field antenna, 2 is the PCB ground plane of the near-field antenna, 31 and 32 are the copper-clad portions of the π-type antenna, 41 and 42 are 50-ohm resistors, 51, 52, and 53 are the ground connection points between the PCB front panel 1 and the PCB ground plane 2; 61 is feed point 1, 62 is feed point 2, 7 is the hole for fixing the plastic part of the PCB front panel 1 and the PCB ground plane 2; and 8 is the hole for fixing the antenna on the PCB ground plane 2.

[0026] Attachment Figure 2 This is the field strength distribution diagram of a common near-field antenna;

[0027] Attachment Figure 3 This is the field strength distribution diagram of the dual-feed point near-field antenna of the present invention;

[0028] Attachment Figure 4 A flowchart of the present invention for establishing a three-dimensional spatial model of an antenna array and outputting the positioning results of the tag in combination with a tag positioning algorithm;

[0029] Attachment Figure 5 This is a flowchart of the positioning result of a single antenna after the antenna array algorithm has obtained the precise location of the existing tag.

[0030] Attachment Figure 6 A three-dimensional spatial model diagram of an antenna array according to a specific embodiment of the present invention;

[0031] Attachment Figure 7 The tag data information table of a specific embodiment of the present invention is ranked by RSSI value to form an RSSI ranking table;

[0032] Attachment Figure 8 This is a flowchart of the positioning algorithm process of a specific embodiment of the present invention. DETAILED DESCRIPTION

[0033] Specific embodiments of the present invention are Figure 1-8 The figure shows an RFID tag positioning system based on a near-field antenna and a tag positioning algorithm, which includes an RFID tag set on an item to be managed, an RFID reader / writer for reading or writing the RFID tag, an antenna array module connected to the RFID reader / writer, a data processing module, and a human-computer interaction module. The above-mentioned data processing module includes a tag positioning algorithm module, an antenna array three-dimensional space generation module, and a tag data information table generation module.

[0034] Arrange the antenna array module in the designated positioning area, partition the designated positioning area into multiple spatial positions, and each antenna in the antenna array corresponds to a spatial position in the positioning area. These spatial positions are numbered in sequence and are called spatial points. x , so each antenna corresponds to a spatial point O x The antenna array module is connected to the antenna splitter of the RFID reader, and the RFID reader is connected to the host computer running the human-computer interaction module and the data processing module. The user operates the system through the human-computer interaction module to locate the items with RFID tags placed in the positioning area. After the RFID reader receives the command to locate the items from the host computer, the antenna splitter on the RFID reader traverses each antenna inventory tag on each antenna array, obtains the RSSI signal value of each tag in the positioning area each time it is inventoried, the number of times each tag is inventoried by each antenna per unit time (such as a patrol cycle), and the antenna number of the inventory tag, and inputs them into the data processing module running on the host computer.

[0035] First, due to the characteristics of RFID technology, the same tag can be read by multiple antennas within a certain range, and the antenna closest to the tag is the actual location of the tag. The positioning algorithm can accurately determine the position of the antenna closest to each tag, but if the field strength distribution of the antenna array is uneven, it will lead to positioning errors. For example, the antenna AntP in the antenna array is located at a nearby spatial point O. x There is a weak area in the space, and the antenna AntQ is at the spatial point O x The middle is the strong area, and the spatial point O xIn space, it belongs to the range of AntP, and the spatial point is O x It is farther from AntQ. When the tag Tagx is located at the spatial point O x When the signal strength and number of times that the AntP antenna reads the tag Tagx are better than the signal strength and number of times that the AntQ antenna reads, the positioning algorithm outputs the positioning result of Tagx as AntQ, resulting in positioning errors.

[0036] Therefore, the antenna used in this embodiment is a dual-feed antenna. The antenna array module is composed of multiple (more than two) traveling wave antennas based on a π-shaped structure. Specifically, the antenna is a 50 ohm characteristic impedance microstrip line with a periodic π-shaped structure. Figure 1 The copper-clad parts 31 and 32 of the π-shaped antenna are shown in the figure; the tail is connected to the 50 ohm resistors 41 and 42. The length of each π-shaped periodic structure is designed to be a quarter wavelength. The distribution of the PCB antenna is a dual antenna distributed on a single PCB board, such as Figure 1 The copper-clad parts 31 and 32 shown in the figure lead to the dual feed points 61 and 62. The dual feed points 61 and 62 of the antenna are connected to the antenna splitter of the RFID reader through feed lines. When the RFID reader is started, feed point 1 (61) and feed point 2 (62) are enabled respectively. When feed point 1 (61) is enabled, the field strength is generated by the copper-clad part 31 of the π-type antenna; when feed point 2 is enabled, the field strength is generated by the copper-clad part 32 of the π-type antenna. The two field strengths complement each other in spatial distribution. Through this π-type structure, the field strength range can be uniform in three-dimensional space, and the distance range is fully controllable, effectively avoiding the blind and weak area problems caused by single-feed point antennas.

[0037] Depend on Figure 2 and Figure 3 The comparison of the field strength distribution diagrams of single feed point and dual feed point shows that Figure 2 The figure shows the field strength distribution diagram of a single feed point. The center area a is the red part, which is the strongest field strength area. From the center to the outside, there are the yellow part of area b (i.e., b1 and b2), the green part of area c, and the blue part of area d. Figure 2 It can be seen that there is a clear boundary between the blue part of area d and the green part of area c. The field strength distribution is relatively strong (areas a, b, and c) and the uniform range is too flat. There are many weak and blind areas on the boundary of the field strength within the coverage range. This will result in a narrow area of ​​stable field strength reading within the entire antenna coverage range, which is prone to missed readings. Figure 3 It can be seen that the central area a is the red part, which is the area with the strongest field strength, and the outermost area d (shown as d1 and d2) is the blue part, which is the area with the weakest field strength. The yellow part of area b and the green part of area c between the central area a and the outermost area d, which are the areas with stronger field strength, are obviously larger than those in the outermost area d. Figure 2 The scope is wide and the Figure 3 The green part of area c in the figure extends with subtle changes in depth, indicating that its field strength is evenly distributed. Overall, the field strength is relatively strong in the central area and has no obvious impact on the surrounding area.

[0038] Secondly, the antenna array three-dimensional space generation module in the data processing module forms a three-dimensional space model of the antenna array according to the spatial point position of each antenna in the antenna array, wherein the antennas are distributed in the spatial point position with their respective feed points as point units. Therefore, in this embodiment, Figure 6 As shown, one spatial point has two antenna numbers (one antenna corresponds to two feed points, so there are two antenna numbers), that is, there are 20 spatial points and 40 antenna numbers. When designing a specific model, the three-dimensional structure can be designed 1:1 using three-dimensional software according to the specific structure of the positioning area. At the same time, the antenna feed point can be positioned in the specific position of the positioning area based on the three-dimensional coordinate system. Alternatively, it can be based on the specific form of the three-dimensional structure. For example, the file cabinet in this embodiment has been partitioned to form 20 spatial positions. In this way, these 20 spatial positions can be numbered and the two antenna numbers distributed therein can be matched, as shown in FIG. Figure 6 In the figure, the spatial point where antenna numbers 1 and 2 (Ant, 1Ant2) are located is O1, and they are distributed in the corresponding order.

[0039] Among them, when the RFID reader receives the command from the host computer to inventory the tags, the RFID reader traverses all the tag information read by each antenna in the antenna array. The tag information includes the RSSI signal value of each tag read by the antenna array, the number of times the tag is read, which antennas read the tag, etc. After the above tag information is input into the tag data information table generation module in the data processing module, a tag data information table is formed. The tag data information table contains the reflected signal value RSSI of each tag each time it is inventoryed. x , the number of times each tag is counted by each antenna per unit time Cnt x And the antenna number Ant that is counted to the label x (That is, the tag data information table contains RSSI x 、Cnt x 、Ant x ). In addition, according to the antenna number Ant x Corresponding to its spatial point (spatial position) in the three-dimensional space model (three-dimensional file cabinet) O X Preferably, adding the spatial point corresponding to the antenna number to the tag data information table is more helpful for the subsequent positioning result calculation; then, the tag data information table is ranked according to the RSSI value to form an RSSI ranking table such as Figure 7 shown.

[0040] Next, the positioning algorithm is combined with the RSSI sorting table of the tag data information table and the advantages and disadvantages of the spatial point position of the tag's antenna number in the three-dimensional spatial model of the antenna array. After integration processing, the optimal antenna spatial point position is output as the positioning position of the tag and transmitted to the host computer.

[0041] Of course, if a tag is only read by one antenna, the spatial point where that antenna number is located can be directly output as the tag location. However, the same tag can often be read by multiple antennas within a certain range, so interference information needs to be eliminated.

[0042] The specific positioning algorithm process is as follows: Under normal circumstances, according to the antenna array design, the distribution pattern of RSSI value and number is: the closer the tag is to the antenna array, the better the RSSI value, and the more times it is read in the same inventory cycle, and vice versa. Therefore, if (1) the RSSI signal value of the tag read by a certain antenna number and the ranking of the number of times the tag is read are both the best in the RSSI sorting table of the tag data information table, it means that the antenna number is closest to the tag, that is, the spatial point where the antenna number in the tag data information table is located is output as the positioning position of the tag; (2) if the RSSI signal value of the tag read by a certain antenna number and the ranking of the number of times the tag is read are not both the best in the RSSI sorting table of the tag data information table, then the two antenna numbers corresponding to the top two RSSI signal values ​​are selected from the RSSI sorting table of the tag data information table, and if the spatial point positions of the two antenna numbers are consistent, then the spatial point positions of the antennas are output as the positioning position of the tag; (3) if the spatial point positions of the two antenna numbers corresponding to the top two RSSI signal values ​​in the RSSI sorting table of the tag data information table are inconsistent, then when one of the RSSI signal values ​​is greater than the other RSSI signal value by a set threshold value ω (the set threshold value ω is initialized to 5dB, and can be subsequently According to the debugging and testing process, and then adjust the initialization assignment based on experience), the spatial point of the antenna number corresponding to the larger RSSI signal value is selected as the positioning position of the tag; (4) If the difference between the first two RSSI signal values ​​in the RSSI sorting table of the tag data information table is less than the set threshold ω, then the two reading times corresponding to the two RSSI signal values ​​are selected for comparison. When the number of readings of one of them is higher than the other by a set threshold σ (the number threshold σ is set to be related to the residence time, that is, when the inventory time / residence time of each antenna in the antenna array is 2000MS, the number threshold σ is set to 3 times, and can be subsequently adjusted based on the debugging and testing process and experience), the spatial point of the antenna number corresponding to the larger number of readings is selected as the positioning position of the tag; (5) If the difference between the two reading times is less than the set threshold σ, the spatial point of the optimal antenna number is output as the positioning position of the tag according to the RSSI signal value, the number of readings and the weighted total score of the spatial point of the antenna number in the two tag data information tables.

[0043] Furthermore, the weighted total score in step ⑤ is calculated according to the weight formula: weight (antenna number spatial point) × coefficient (antenna number spatial point) + weight (RSSI) × coefficient (RSSI) + weight (number of times read) × coefficient (number of times read) = total score, where weight (antenna number spatial point) > weight (RSSI) > weight (number of times read). The two tag data information tables are sorted by the total scores calculated according to the weight formula, and the spatial point of the antenna number in the tag data information table with the highest total score is selected as the tag positioning position. For example: Tag Tag1 is detected by three antennas Ant at the same time in a complete cycle. x ,Ant y ,Ant z After reading, sorted by RSSI value, they are (Ant x , RSSI x =-64.5, Cnt x =8, the spatial point is 15), (Ant y , RSSI y =-66.2, Cnt y =7, the spatial point is 16), (Ant z , RSSI z =-66.9, Cnt z =9, area number 13). The first-place ranking in each category receives 3 points, the second-place ranking receives 2 points, and the third-place ranking receives 1 point. Weighting coefficients: RSSI value is 0.3, number of times is 0.2, and position is 0.4. The scores and weighting coefficients can be set manually or empirically; however, a more reliable approach is to train the BP neural network using a large number of data samples to determine these values. During actual debugging and testing, N sets of tests can be performed to select data with similar RSSI values ​​and number of times. The coefficients can then be inferred based on their true values. Multiple sets of data requiring weight calculation and their true results are selected and trained through the BP neural network to determine the scores and weighting coefficients. Neural network training techniques are common practice, so the training process will not be detailed here. Since the positioning algorithm described above reaches the point where differentiation based on the total weighted score is required, a direct result cannot be obtained by comparing only RSSI values ​​and number of times. This comparison is based on the small differences in RSSI values ​​and number of times. Therefore, the highest weighting coefficient is assigned to position, followed by RSSI value and number of times. The position is determined by the distance from the middle position. Figure 6 It can be seen that spatial point 15 is closest to the center of the filing cabinet, followed by spatial point 16, and spatial point 13 is the last. Then:

[0044] Antx=3×0.3+2×0.2+3×0.4=2.5

[0045] Anty=2×0.3+1×0.2+2×0.4=1.6

[0046] Antz=1×0.3+3×0.2+1×0.4=1.3

[0047] Therefore, the optimal output result is that the area where Antx is located is 15.

[0048] According to the above-mentioned three-dimensional spatial model of the antenna array and the tag positioning algorithm, the precise location of the existing tag can be obtained. Since the antenna array has a wide range of uses, it can be used in the inventory of the entire cabinet, and can also be used in smart cabinets with multiple compartments to locate tags or objects. The difference between the two is that when the entire cabinet is inventoried, all antenna arrays need to be polled, so it takes a long time; and when a compartment of a smart cabinet with multiple compartments is opened separately, only the dual-feed point antenna corresponding to the compartment can be triggered for inventory, so the time period can be relatively short. When a single antenna is triggered to read the tag, if the tag is increased or decreased, it is only necessary to design a positioning logic algorithm based on the tag information read by the single antenna and the data information library of the original antenna array to obtain the increase or decrease of the tag. Specific description:

[0049] (1) The antenna numbers in the antenna array are distributed as Ant1, Ant2, Ant3...Antn

[0050] (2) The label data in the antenna array database is defined as:

[0051] Ant1 antenna number data: Ant1-Tag1, Ant1- Tag2,... Ant1- Tag m1

[0052] Ant2 antenna number data: Ant2- Tag1, Ant2- Tag2,... Ant2- Tag m2...

[0053] Antn antenna number data: Antn- Tag1, Antn- Tag2,... Antn- Tag mn

[0054] (3) During a tag data reading cycle of an antenna Antx in the antenna array, w tag data are read, named R1, R2, ..., Rw.

[0055] The first step is to determine whether Ri is in the Antx information library (Antx- Tag1, Antx- Tag2,... Antx-Tag m1 ):

[0056] If Ri is in the Antx repository, then Ri => status is unchanged;

[0057] If Ri is not in the Antx information library, check whether Ri is in all antenna array databases (Ant1-Antn). If it is in the database, it means that Ri is a tag in other antennas read in series, and Ri=>status remains unchanged; if Ri is not in the entire database, Ri=>is added;

[0058] Polling starts from R1 and ends at Rw.

[0059] The second step is to extract the tag data from the Antx database and record it as: Antx- Tag1, Antx-Tag2, Antx- Tag k,..., Antx- Tag mx;

[0060] Search Antx- Tag1, Antx- Tag2,..., Antx- Tag in R1-R30 in turn k,..., Antx-Tag mx ,

[0061] If you can find Antx- Tag in R1~R30 k , then Antx- Tag k =>The state has not changed;

[0062] If Antx- Tag is not found in R1~R30 k , then Antx- Tag k =>Take out;

[0063] Through the above comparative analysis, it can be concluded that after a single antenna reads the tags, it can accurately locate the newly added and removed tags.

[0064] Specific experimental effect verification: Both the experimental group and the control group used an intelligent file positioning cabinet (length × width × height = 1000mm × 360×2000mm) as the designated area, which was divided into 5 layers and 20 area grids, and a near-field antenna was arranged in each area grid.

[0065] The control group used a common near-field antenna in a scenario where the antenna array locates the tag position. There were 20 grid areas, with a total of 20 near-field antenna arrays. RFID files were placed in each area in sequence, with the number of files gradually increasing. Each group of samples was tested 500 times. The test results are as follows:

[0066] When an average of 20 RFID documents are placed in each area and the entire cabinet contains 400 RFID documents, the probability of cross-reading is 1% and the probability of missed reading is 1.5%.

[0067] When an average of 30 RFID documents are placed in each area and the entire cabinet contains 600 RFID documents, the probability of cross-reading is 3% and the probability of missed reading is 2%.

[0068] When an average of 40 RFID documents are placed in each area and the entire cabinet contains 800 RFID documents, the probability of cross-reading is 4.5% and the probability of missed reading is 3%.

[0069] The positioning results of 1,000 RFID documents were tested 500 times in a row. The probability of cross-reading was more than 5%, and the probability of missed reading was more than 3%.

[0070] In the test data above, as the number of RFID files increases, the overlap of tags increases, the gaps between adjacent tags decrease, and the tag group reading effect becomes worse. RFID tags covered by the weak area of ​​the antenna array may not be read, resulting in missed reading in this area, while the tag can be read by other antenna arrays in adjacent areas, thus resulting in a false positive result of cross-reading.

[0071] The experimental group used the aforementioned dual-feed near-field antenna method and combined it with a positioning algorithm. Instead of the original 20 near-field antenna arrays, the arrays were expanded to 40 within the same spatial distribution. 20 to 50 RFID documents were placed in each area for comparative testing. Each group of samples underwent 500 tests.

[0072] When an average of 20 RFID files are placed in each area and the entire cabinet contains 400 RFID files, the probability of cross-reading is 0% and the probability of missed reading is 0%.

[0073] When an average of 30 RFID documents are placed in each area and the entire cabinet contains 600 RFID documents, the probability of cross-reading is 0% and the probability of missed reading is 0%.

[0074] When an average of 40 RFID documents are placed in each area and the entire cabinet contains 800 RFID documents, the probability of cross-reading is 0% and the probability of missed reading is 0.1%.

[0075] When an average of 50 RFID documents are placed in each area, the positioning results of a total of 1,000 RFID documents in the cabinet are tested. The probability of cross-reading is less than 1%; the probability of missed reading is less than 0.5%.

[0076] It can be seen from this that when the antenna array formed by the above antenna group is used in application scenarios such as file positioning / object positioning, the uniformity of the field strength distribution of each antenna in the local area can be strictly controlled. Based on this antenna array, the tag signal strength read by the RFID reader is evenly distributed. Combined with the positioning algorithm, it can avoid missed reads and obvious cross-reading between antenna arrays in adjacent areas.

[0077] The present invention is not limited to the above-mentioned specific embodiments. Those skilled in the art can implement the present invention in various other specific embodiments based on the contents disclosed in the present invention, or any simple changes or modifications made to the design structure and concept of the present invention fall within the scope of protection of the present invention.

Claims

1. An RFID tag positioning system based on a near-field antenna and a tag positioning algorithm, comprising an RFID tag attached to an item to be managed, an RFID reader / writer for reading or writing the RFID tag, an antenna array module connected to the RFID reader / writer, a data processing module, and a human-computer interaction module, characterized in that: The data processing module includes a tag positioning algorithm module, an antenna array three-dimensional space generation module and a tag data information table generation module. Arrange the antenna array module in the designated positioning area, partition the designated positioning area into multiple spatial positions, and number these spatial positions in sequence, called spatial points. Each antenna corresponds to a spatial point. Connect each antenna array to the antenna splitter of the RFID reader, connect the antenna array module to the RFID reader, and connect the RFID reader to the host computer running the human-computer interaction module and the data processing module. The user operates the system through the human-computer interaction module to locate the items with RFID tags placed in the positioning area. After the RFID reader receives the command from the host computer to locate the item, it traverses each antenna on each antenna array to count the tags, obtains the RSSI signal value of each tag in the positioning area when it is counted, the number of times each tag is counted by each antenna per unit time, and the antenna number of the tag, and inputs it into the data processing module running on the host computer. The antenna array three-dimensional space generation module in the data processing module forms a three-dimensional space model of the antenna array according to the spatial point position of each antenna in the antenna array, wherein the antennas are distributed in the spatial point position based on the antenna feed point, and the closer the spatial point position of the antenna is to the center of the three-dimensional space model, the better its position. The tag data information table generation module in the data processing module generates a tag data information table for each tag based on the RSSI signal value each time the tag is counted, the number of times each tag is counted by each antenna per unit time, and the antenna number of the tag. The RSSI signal values ​​in the tag data information table are sorted and calculated according to their strength to obtain an RSSI sorting table of the tag data information table. The tag positioning algorithm module in the data processing module integrates the RSSI sorting table of the tag data information table and the advantages and disadvantages of the spatial point position of the tag's antenna number in the three-dimensional spatial model of the antenna array, and then outputs the optimal antenna spatial point position as the positioning position of the tag and transmits it to the host computer.

2. The RFID tag positioning system based on near-field antenna and tag positioning algorithm according to claim 1, characterized in that: The antenna array module is composed of a plurality of traveling wave antennas based on a π-type structure, and the traveling wave antenna includes a dual-feed point structure.

3. The RFID tag positioning system based on near-field antenna and tag positioning algorithm according to claim 2, characterized in that: The traveling wave antenna includes a PCB board, on which two groups of π-shaped structure traces are distributed, one upper group and one lower group. A feeding point is independently provided at one end of each group of π-shaped structure traces, and the two feeding points are distributed one on the left and one on the right. The two groups of π-shaped structure waveforms complement each other in spatial distribution.

4. An RFID tag positioning method based on the RFID tag positioning system according to any one of claims 1 to 3, characterized in that: The following steps are involved: ① Arrange antenna array modules in a designated positioning area, partition the designated positioning area into multiple spatial locations, and number these spatial locations sequentially, called spatial points. Each antenna corresponds to a spatial point, and each antenna array is connected to the antenna splitter of the RFID reader. ② After receiving a command from the host computer to locate and count items with RFID tags in the positioning area, the RFID reader completes the data inventory of the tags in the positioning area, obtains the RSSI signal value of each tag in the positioning area when it is counted, the number of times the tag is counted by each antenna per unit time, and the antenna number of each tag counted, and inputs this information into the data processing module running on the host computer; ③ The spatial position of each antenna in the antenna array is input into the antenna array three-dimensional space generation module in the data processing module to form a three-dimensional space model of the antenna array, wherein the antennas are distributed in the spatial position with the antenna feed point as the unit, and the closer the spatial position of the antenna is to the center of the three-dimensional space model, the better its position is; ④ The RSSI signal value of each tag each time it is inventoried, the number of times each tag is inventoried by each antenna per unit time, and the antenna number of the inventory tag are input into the tag data information table generation module in the data processing module to generate a tag data information table for each tag, and after sorting and calculating according to the strength of the RSSI signal value in the tag data information table, the RSSI sorting table of the tag data information table is obtained; ⑤ The tag positioning algorithm module in the data processing module performs tag positioning processing based on the advantages and disadvantages of the spatial point positions of the antennas in the three-dimensional space model of the antenna array and the RSSI sorting table of the tag data information table, and then outputs the optimal antenna spatial point position as the positioning position of the tag and transmits it to the host computer.

5. The RFID tag positioning method based on near-field antenna and tag positioning algorithm according to claim 4, characterized in that: In step ⑤, the tag positioning algorithm processing includes the following process: (1) if there is a tag data information table in which the RSSI signal value of the tag read by a certain antenna number and the ranking of the number of times the tag is read are both optimal, then the spatial point where the antenna number in the tag data information table is located is output as the positioning position of the tag; (2) If the RSSI signal value of a tag read by a certain antenna number and the number of times the tag is read are not optimal in the RSSI sorting table of the tag data information table, the optimal spatial point of the antenna number in the tag data information table is output as the positioning position of the tag according to the total weight of the RSSI signal value, the number of times the tag is read and the spatial point of the antenna number in the tag data information table.

6. The RFID tag positioning method based on near-field antenna and tag positioning algorithm according to claim 5, characterized in that: In step ⑤, if there is a situation where the RSSI signal value of the tag read by a certain antenna number exists in the RSSI sorting table of the tag data information table and the ranking of the number of times the tag is read is not optimal, then the two antenna numbers corresponding to the first two RSSI signal values ​​are selected from the RSSI sorting table of the tag data information table, and if the spatial point positions of the two antenna numbers are consistent, the spatial point positions of the antenna are output as the positioning position of the tag; if the spatial point positions of the two antenna numbers corresponding to the first two RSSI signal values ​​in the RSSI sorting table of the tag data information table are inconsistent, then when one of the RSSI signal values ​​is more than the set threshold value ω than the other RSSI signal value, the RSSI signal value with the larger value is selected. The corresponding antenna number spatial point is used as the positioning position of the tag; if the difference between the first two RSSI signal values ​​in the RSSI sorting table of the tag data information table is less than the set threshold ω, the two reading times corresponding to the two RSSI signal values ​​are selected for comparison. When the number of readings of one of them is higher than the other by a set threshold σ, the spatial point of the antenna number corresponding to the larger number of readings is selected as the positioning position of the tag; if the difference between the two reading times is less than the set threshold σ, the spatial point of the optimal antenna number is output as the positioning position of the tag according to the RSSI signal value, the number of readings and the weighted total score of the spatial point of the antenna number in the two tag data information tables.

7. The RFID tag positioning method based on near-field antenna and tag positioning algorithm according to claim 6, characterized in that: The weighted total score in step ⑤ is calculated according to the weight formula: weight (antenna number spatial point) × coefficient (antenna number spatial point) + weight (RSSI) × coefficient (RSSI) + weight (number of times read) × coefficient (number of times read) = total score, where weight (antenna number spatial point) > weight (RSSI) > weight (number of times read). The two tag data information tables are sorted by the total scores calculated according to the weight formula, and the spatial point where the antenna number in the tag data information table with the highest total score is located is selected as the tag positioning position.

8. The RFID tag positioning method based on near-field antenna and tag positioning algorithm according to claim 4, characterized in that: The antenna array adopts a traveling wave antenna based on a π-type structure. The traveling wave antenna includes a PCB board, on which two groups of π-type structure traces are distributed, one upper and one lower. Each end of the two groups of π-type structure traces is independently provided with a feeding point, and the two feeding points are distributed one on the left and one on the right. The two groups of π-type structure waveforms complement each other in spatial distribution.

9. The RFID tag positioning method based on near-field antenna and tag positioning algorithm according to claim 8, characterized in that: The PCB board includes a PCB front panel and a PCB ground panel. The PCB front panel is distributed with two antenna copper clad parts with corresponding π-shaped structure routing. A feeding point is provided at one end of each of the two antenna copper clad parts, and the two feeding points are distributed on the left and right sides.

Images