A vehicle electronic license plate recognition method and system

Abstract

This invention discloses a vehicle electronic tag identification method and system. The method includes: a reader sending a query command specifying a frame length; tags within the reader's identification range randomly selecting a time slot within the frame length to respond to the reader's command and return an information packet; if a collision time slot exists, acquiring the observation values ​​from the previous frame; obtaining the number of tags based on the observation values ​​from the previous frame and a preset tag estimation algorithm; obtaining the number of time slots in the next frame based on the number of tags and the number of antennas; if the number of antennas is not greater than the number of tags in the next frame's time slots, then using a second preset separation method to separate the tag signals, and performing vehicle identification based on the separated signals. By selecting a suitable separation method, separating the signals, and completing the identification, the accuracy of identification is improved, identification errors caused by collisions are avoided, and identification efficiency is increased.

Description

Technical Field

[0001] This application relates to the field of vehicle identification technology, and more specifically, to a method and system for identifying electronic vehicle identification tags. Background Technology

[0002] Radio Frequency Identification (RFID) is an automatic identification technology that uses wireless radio frequency (RF) for non-contact, two-way data communication. It reads and writes data to a recording medium (electronic tag or RFID card) to identify targets and exchange data. It is considered one of the most promising information technologies of the 21st century. An RFID system typically consists of at least two parts: an electronic tag and a reader. The electronic tag usually stores electronic data in a pre-defined format and is attached to the surface of the object to be identified. The reader, also known as a reading device, reads and identifies the electronic data stored in the tag without contact, thus achieving automatic object identification. Further management functions, such as the collection, processing, and remote transmission of object identification information, are achieved through computers and computer networks.

[0003] Multi-antenna technology is not simply about increasing the number of antennas, but rather refers to smart antennas with algorithms for tracking signals and locating signal sources. Compared with single-antenna technology, multi-antenna technology has obvious advantages, including not only large communication capacity, fast transmission rate, low signal transmission power, and the ability to locate signal sources, but also increased system capacity, improved spectrum efficiency, and expanded signal coverage.

[0004] Currently, many scholars have conducted research on multi-antenna RFID systems and developed a 4-antenna UHF RFID reader that is unaffected by tag orientation when reading tags, enabling the instantaneous reading of multiple tags and increasing the number of tags that can be read. However, the problem of low identification efficiency still exists in scenarios with a large number of tags, such as logistics management, warehouse management, and vehicle management. In multi-antenna RFID systems, a large number of tags transmit signals to multiple antennas of the reader within the same time period. This increases the probability of tag signal collisions at a certain point in time, thus affecting the identification efficiency of the multi-antenna RFID system.

[0005] A diagram illustrating tag collision is shown below. Figure 3 As shown, tag collision refers to the situation where two or more tag signals simultaneously send signals back to the reader at the same time. As a result, the signals returned at the same time will collide with each other, causing the colliding tags to become unidentifiable. Therefore, tag collision reduces the identification efficiency of the RFID system.

[0006] Therefore, how to improve the efficiency of label recognition is a technical problem that needs to be solved. Summary of the Invention

[0007] This invention provides a vehicle electronic tag identification method to solve the technical problem of low tag identification efficiency in existing technologies. The method is applied to a multi-antenna RFID system, which has a preset number of antennas, and includes:

[0008] Step 1: The reader sends a query command specifying the frame length. Tags within the reader's recognition range randomly select a time slot within the frame length range to respond to the reader's command and return an information packet.

[0009] Step 2: If a collision time slot exists, after the previous frame ends, obtain the observation value in the previous frame, and obtain the number of labels based on the observation value in the previous frame and the preset label estimation algorithm;

[0010] Step 3: Based on the number of tags and the number of antennas, obtain the number of time slots for the next frame. Based on the number of time slots for the next frame, determine whether the number of antennas is greater than the number of tags in the time slots of the next frame.

[0011] Step 4: If the number of antennas is greater than the number of tags in the next frame time slot, the first preset separation method is used to separate the tag signals, and vehicle identification is performed based on the separated signals. If the number of antennas is not greater than the number of tags in the next frame time slot, the second preset separation method is used to separate the tag signals, and vehicle identification is performed based on the separated signals.

[0012] The collision time slot refers to the time slot of two or more tag return information packets.

[0013] In some embodiments of this application, the number of tags is obtained based on the observations in the previous frame and a preset tag estimation algorithm, specifically as follows:

[0014] The observed values ​​include the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of idle time slots is the number of time slots in which no tag returns a packet. The number of successful time slots is the number of time slots in which only one tag returns a packet. The number of collision time slots is the number of time slots in which two or more tags return packets.

[0015] The probability of a collision slot appearing in the initial frame is calculated based on the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of tags is then roughly estimated based on the probability to obtain the first number of tags. If the first number of tags does not meet the preset requirements, a fine estimate is performed to obtain the number of tags.

[0016] If the number of the first tag meets the preset requirement, then the number of the first tag will be used as the total number of tags.

[0017] In some embodiments of this application, if the number of antennas is not greater than the number of tags in the next frame time slot, a second preset separation method is used to separate the tag signals, specifically:

[0018] Let there be N tags in a time slot, and the return signals of the N tags be S = [S1, S2, S3, ..., S2]. N The reader has M antennas, which receive M mixed signals X = [X1, X2, X3...X...] from N tags in a random manner. M ], where X = AS, A is an N×M dimensional instantaneous linear mixture matrix with full column rank, and the separation matrix W is obtained by ICA method. Substituting the separation matrix into the following model:

[0019] Y = WX,

[0020] Where Y is the signal that is close to the tag signal after separation of M mixed signals, and W is an M×N dimensional instantaneous linear separation matrix.

[0021] In some embodiments of this application, the method further includes:

[0022] If the signal separated by the first or second preset separation method does not contain any unidentified tags, then vehicle identification is performed based on the separated signal.

[0023] If the signal separated by the first or second preset separation method contains unidentified tags, then repeat steps one through four.

[0024] Correspondingly, this application also provides a vehicle electronic identification system, applied to a multi-antenna RFID system, which has a preset number of antennas, and the system includes:

[0025] The response module is used by the reader to send a query command specifying the frame length. Tags within the reader's recognition range randomly select a time slot within the frame length range to respond to the reader's command and return an information packet.

[0026] The estimation module is used to obtain the observation value in the previous frame after the previous frame ends if a collision time slot exists, and obtain the number of tags based on the observation value in the previous frame and a preset tag estimation algorithm.

[0027] The judgment module is used to obtain the number of time slots in the next frame based on the number of tags and the number of antennas, and to determine whether the number of antennas is greater than the number of tags in the time slots of the next frame based on the number of time slots in the next frame.

[0028] The separation module is used to separate the tag signal using a first preset separation method if the number of antennas is greater than the number of tags in the next frame time slot, and to identify the vehicle based on the separated signal; if the number of antennas is not greater than the number of tags in the next frame time slot, it is used to separate the tag signal using a second preset separation method, and to identify the vehicle based on the separated signal.

[0029] The collision time slot refers to the time slot of two or more tag return information packets.

[0030] In some embodiments of this application, the estimation module is specifically used for:

[0031] The observed values ​​include the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of idle time slots is the number of time slots in which no tag returns a packet. The number of successful time slots is the number of time slots in which only one tag returns a packet. The number of collision time slots is the number of time slots in which two or more tags return packets.

[0032] The probability of a collision slot appearing in the initial frame is calculated based on the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of tags is then roughly estimated based on the probability to obtain the first number of tags. If the first number of tags does not meet the preset requirements, a fine estimate is performed to obtain the number of tags.

[0033] If the number of the first tag meets the preset requirement, then the number of the first tag will be used as the total number of tags.

[0034] In some embodiments of this application, the separation module is specifically used for:

[0035] Let there be N tags in a time slot, and the return signals of the N tags be S = [S1, S2, S3, ..., S2]. N The reader has M antennas, which receive M mixed signals X = [X1, X2, X3...X...] from N tags in a random manner. M ], where X = AS, A is an N×M dimensional instantaneous linear mixture matrix with full column rank, and the separation matrix W is obtained by ICA method. Substituting the separation matrix into the following model:

[0036] Y = WX,

[0037] Where Y is the signal that is close to the tag signal after separation of M mixed signals, and W is an M×N dimensional instantaneous linear separation matrix.

[0038] In some embodiments of this application, the system further includes a verification module, which is used for:

[0039] If the signal separated by the first or second preset separation method does not contain any unidentified tags, then vehicle identification is performed based on the separated signal.

[0040] If the signal separated by the first or second preset separation method contains unidentified tags, then repeat steps one through four.

[0041] By applying the above technical solution, the process is as follows: Step 1: The reader sends a query command specifying the frame length. Tags within the reader's recognition range randomly select a time slot within the frame length to respond to the reader's command and return an information packet. Step 2: If a collision time slot exists, after the previous frame ends, the observation values ​​from the previous frame are obtained. The number of tags is obtained based on the observation values ​​from the previous frame and a preset tag estimation algorithm. Step 3: The number of time slots in the next frame is obtained based on the number of tags and the number of antennas. Based on the number of time slots in the next frame, it is determined whether the number of antennas is greater than the number of tags in the next frame's time slots. Step 4: If the number of antennas is greater than the number of tags in the next frame's time slots, a first preset separation method is used to separate the tag signals, and vehicle identification is performed based on the separated signals. If the number of antennas is not greater than the number of tags in the next frame's time slots, a second preset separation method is used to separate the tag signals, and vehicle identification is performed based on the separated signals. This application obtains the estimated frame length of the next frame based on the number of antennas and tags, and selects an appropriate separation method based on the relationship between the number of antennas and the number of tags in the time slots to separate the signals and complete the identification. It improves the accuracy of recognition, avoids recognition errors caused by collisions, and improves recognition efficiency. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 A flowchart illustrating a vehicle electronic identification method according to an embodiment of the present invention is shown;

[0044] Figure 2 A schematic diagram of the structure of a vehicle electronic identification system proposed in an embodiment of the present invention is shown;

[0045] Figure 3 A schematic diagram illustrating the tag collision principle in the background art of this invention is shown;

[0046] Figure 4 A schematic diagram of a mathematical model for blind origin separation proposed in another embodiment of the present invention is shown;

[0047] Figure 5 A schematic diagram of a mathematical model based on the ICA method proposed in another embodiment of the present invention is shown. Detailed Implementation

[0048] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] This application provides a vehicle electronic identification method, applied to a multi-antenna RFID system, wherein the number of antennas is preset, such as... Figure 1 As shown, the method includes the following steps:

[0050] Step 1: S101, the reader sends a Query command specifying the frame length. Tags within the reader's recognition range randomly select a time slot within the frame length range to respond to the reader's command and return an information packet.

[0051] In this embodiment, the slotted ALOHA algorithm is improved. The basic workflow of the slotted ALOHA algorithm is as follows: the reader first sends a Query command specifying the frame length, i.e., the number of frame slots; tags within the identification range randomly select a slot within the frame length to respond to the reader's command and return an information packet. A slot where only one tag returns an information packet is called a successful slot; a slot where no tag returns an information packet is called an empty slot; and a slot where two or more tags return information packets is called a collision slot. Tags that collide will continue to attempt identification in the next frame. The algorithm first uses a certain tag estimation method to estimate the number of tags in the field area based on the feedback of the previous frame (i.e., the observation values: the number of collision slots, the number of empty slots, and the number of successful slots). Then, based on the number of tags and other conditions, a frame slot estimation algorithm is used to estimate the frame length of the next frame, and so on, until all tags in the reader's working area have been identified. The improvement is that the frame length is changed according to the current number of tags and the number of antennas, thereby selecting an appropriate number of slots.

[0052] Step 2: S102, if a collision time slot exists, after the previous frame ends, obtain the observation value in the previous frame, and obtain the number of tags based on the observation value in the previous frame and the preset tag estimation algorithm.

[0053] In some embodiments of this application, the number of tags is obtained based on the observations in the previous frame and a preset tag estimation algorithm. Specifically, the observations include the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of idle time slots is the number of time slots with no tag return packets, the number of successful time slots is the number of time slots with only one tag return packet, and the number of collision time slots is the number of time slots with two or more tag return packets. The probability of a collision time slot appearing in the initial frame is calculated based on the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of tags is coarsely estimated based on the probability to obtain a first number of tags. If the first number of tags does not meet the preset requirements, a fine estimation is performed to obtain the number of tags. If the first number of tags meets the preset requirements, the first number of tags is used as the total number of tags.

[0054] In this embodiment, the initial frame size is first set to F. After the recognition of the previous frame ends, the number of idle time slots C0, the number of successful time slots C1, and the number of collision time slots CK are counted. Then, the probability P = CK / F of a collision time slot appearing in the initial frame is calculated. The relationship between P and the average number of tags nk in the collision time slots is analyzed to estimate the number of tags N (the first number of tags) in the first calculation. Then, the estimated number of tags in the first calculation is evaluated. If it meets the requirements, the algorithm ends; otherwise, a fine estimation, i.e., a second tag number estimation, is required. The fine estimation algorithm adopts the maximum a posteriori probability estimation algorithm based on prior knowledge. The number of tags N estimated in the coarse estimation is used as the starting value for the fine estimation. The search direction in the coarse estimation is determined by the posterior probability. Then, the maximum a posteriori probability search is performed until the requirements are met and the search stops. The N at this point is the final estimated value of the number of tags.

[0055] like Figure 3 As shown, blind source separation refers to the process of determining the source signal data solely by obtaining the observed signal data, based on certain statistical characteristics of the source signal, when the parameters of the source signal and the transmission channel are unknown. Applying the concept of blind source separation to the processing of tag signals in a multi-antenna RFID system, the mathematical model of blind source separation is similar to the mathematical model of multiple antennas receiving multiple tag signals in an RFID system. The left side of the diagram represents M source signals S, analogous to multiple tag signals in a certain time slot in a multi-antenna RFID system. A represents the mixing matrix, both of which are unknown. The middle section represents N observed signals X, analogous to signals received by multiple antennas in a certain time slot in a multi-antenna RFID system. The observed signals are known and have the same length. The mathematical model for blind source separation can be expressed as the formula:

[0056] X = AS, Y = WX.

[0057] Y is the output signal after passing through the separation matrix. The blind source separation method aims to find a separation matrix w that makes the output signal approximate the true source signal as closely as possible. The mixing matrix A is an N×M dimensional instantaneous linear mixing matrix with full column rank; the separation matrix W is an M×N dimensional instantaneous linear separation matrix.

[0058] Step 3: S103, based on the number of tags and the number of antennas, obtain the number of time slots for the next frame, and based on the number of time slots for the next frame, determine whether the number of antennas is greater than the number of tags in the time slots of the next frame.

[0059] In this embodiment, the number of tags in the next frame time slot is the number of tags to be identified. When the first frame ends, the number of empty time slots, successful time slots, and failed time slots in this frame is known. The total number of tags, n, can be estimated using existing tag estimation algorithms. Then, the number of tags to be identified in the next frame can be calculated using n and the number of tags already identified in this frame. The optimal number of time slots for the next frame can be estimated. The number of time slots N1 in the next frame is determined by the number of antennas, the number of tags, and the number of time slots in the initial frame. It is determined whether the number of antennas is greater than the number of tags in the next frame time slot. The number of tags in the next frame time slot is N, and the number of antennas is M. The magnitudes of M and N are compared, and a separation method is selected based on their relationship.

[0060] Step 4: S104, if the number of antennas is greater than the number of tags in the next frame time slot, the first preset separation method is used to separate the tag signals, and vehicle identification is performed based on the separated signals. If the number of antennas is not greater than the number of tags in the next frame time slot, the second preset separation method is used to separate the tag signals, and vehicle identification is performed based on the separated signals.

[0061] In some embodiments of this application, if the number of antennas is not greater than the number of tags in the next frame time slot, a second preset separation method is used to separate the tag signals. Specifically, this involves setting a time slot to contain N tags, and the return signals of the N tags to be S = [S1, S2, S3...S...]. N The reader has M antennas, which receive M mixed signals X = [X1, X2, X3...X...] from N tags in a random manner. M ], where X = AS, A is an N×M dimensional instantaneous linear mixing matrix with full column rank, and the separation matrix W is obtained by ICA method. Substitute the separation matrix into the following model: Y = WX, where Y is the signal that is close to the tag signal after separation of M mixed signals, and W is an M×N dimensional instantaneous linear separation matrix.

[0062] In this embodiment, the ICA method is based on independent component analysis, and the number of observed signals is equal to or greater than the number of source signals. These two cases can be regarded as one case. In blind source separation, at this time, the mixing matrix A is a square matrix, the A matrix is invertible, and the unique solution of the source signal can be obtained.

[0063] As Figure 4 shown, there are N tags in a time slot, and the signals returned by the N tags are denoted as S = [S1, S2, S3......S N , which are source signals. The reader has M antennas (usually no more than 8 affected by other factors). The M antennas receive M mixed signals that are irregularly mixed by the N tag signals, denoted as X = [X1, X2, X3......X M , which are the observed signals. After being processed by the ICA method of the reader, these M mixed signals are separated into signals close to the tag signals, denoted as Y = [Y1, Y2, Y3......Y N . In the ICA method, the number of observed signals is less than or equal to the number of source signals, that is, M ≤ N. The mathematical model of blind source separation is X = AS. Using the ICA method, the separation matrix W can be obtained, so that the mixed signal X received by the antenna outputs Y = [Y1, Y2, Y3......Y N which is the optimal approximation of the source signal S. That is, Y = WX. The key to the idea of blind source separation is to obtain the separation matrix W, and then separate the received mixed signal through matrix transformation to obtain the separation signal Y that is closest to the source signal X, and the tag signals in collision in the time slot are successfully identified. However, the use of the ICA method is also related to the number of time slots in this frame. Only the optimal number of time slots can better improve the recognition efficiency of the system. If the number of time slots is too small, the number of tags in each time slot will be more. When the number of tags in the time slot is more than the number of antennas, the ICA method cannot be used to identify the collision tags; if the number of time slots is too large, the number of idle time slots will be more, wasting system time. Therefore, the size of the number of time slots will affect the recognition efficiency of the system.

[0064] When M > N, that is, the number of observed signals is greater than the number of source signals (N < M); at this time, the number of columns in the mixing matrix A is more than the number of rows, the matrix is irreversible, and there is no unique solution of the source signal. The ICA method cannot be used to separate the signals. Generally, the two-step method of sparse component analysis (SCA) is used to separate the signals. This technology is a conventional technical means of this field and will not be elaborated here.

[0065] In some embodiments of this application, the method further includes: if the signal separated by the first preset separation method or the second preset separation method does not contain an unidentified tag, then vehicle identification is performed based on the separated signal; if the signal separated by the first preset separation method or the second preset separation method contains an unidentified tag, then steps one to four are repeated.

[0066] By applying the above technical solution, the process is as follows: Step 1: The reader sends a query command specifying the frame length. Tags within the reader's recognition range randomly select a time slot within the frame length to respond to the reader's command and return an information packet. Step 2: If a collision time slot exists, after the previous frame ends, the observation values ​​from the previous frame are obtained. The number of tags is obtained based on the observation values ​​from the previous frame and a preset tag estimation algorithm. Step 3: The number of time slots in the next frame is obtained based on the number of tags and the number of antennas. Based on the number of time slots in the next frame, it is determined whether the number of antennas is greater than the number of tags in the next frame's time slots. Step 4: If the number of antennas is greater than the number of tags in the next frame's time slots, a first preset separation method is used to separate the tag signals, and vehicle identification is performed based on the separated signals. If the number of antennas is not greater than the number of tags in the next frame's time slots, a second preset separation method is used to separate the tag signals, and vehicle identification is performed based on the separated signals. This application obtains the estimated frame length of the next frame based on the number of antennas and tags, and selects an appropriate separation method based on the relationship between the number of antennas and the number of tags in the time slots to separate the signals and complete the identification. It improves the accuracy of recognition, avoids recognition errors caused by collisions, and improves recognition efficiency.

[0067] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0068] To further illustrate the technical concept of this invention, the technical solution of this invention will now be described in conjunction with specific application scenarios.

[0069] Correspondingly, this application also provides a vehicle electronic identification system, applied to a multi-antenna RFID system, which has a preset number of antennas, and the system includes:

[0070] Response module 201 is used for the reader to send a query command specifying the frame length, and for tags within the reader's recognition range to randomly select a time slot within the frame length range to respond to the reader's command and return an information packet;

[0071] The estimation module 202 is used to obtain the observation value in the previous frame after the previous frame ends if a collision time slot exists, and obtain the number of tags based on the observation value in the previous frame and a preset tag estimation algorithm.

[0072] The judgment module 203 is used to obtain the number of time slots in the next frame based on the number of tags and the number of antennas, and to determine whether the number of antennas is greater than the number of tags in the time slots of the next frame based on the number of time slots in the next frame.

[0073] The separation module 204 is used to separate the tag signal using a first preset separation method if the number of antennas is greater than the number of tags in the next frame time slot, and to identify the vehicle based on the separated signal; if the number of antennas is not greater than the number of tags in the next frame time slot, it is used to separate the tag signal using a second preset separation method, and to identify the vehicle based on the separated signal.

[0074] The collision time slot refers to the time slot of two or more tag return information packets.

[0075] In some embodiments of this application, the estimation module 202 is specifically used for:

[0076] The observed values ​​include the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of idle time slots is the number of time slots in which no tag returns a packet. The number of successful time slots is the number of time slots in which only one tag returns a packet. The number of collision time slots is the number of time slots in which two or more tags return packets.

[0077] The probability of a collision slot appearing in the initial frame is calculated based on the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of tags is then roughly estimated based on the probability to obtain the first number of tags. If the first number of tags does not meet the preset requirements, a fine estimate is performed to obtain the number of tags.

[0078] If the number of the first tag meets the preset requirement, then the number of the first tag will be used as the total number of tags.

[0079] In some embodiments of this application, the separation module 204 is specifically used for:

[0080] Let there be N tags in a time slot, and the return signals of the N tags be S = [S1, S2, S3, ..., S2]. N The reader has M antennas, which receive M mixed signals X = [X1, X2, X3...X...] from N tags in a random manner.M ], where X = AS, A is an N×M dimensional instantaneous linear mixture matrix with full column rank, and the separation matrix W is obtained by ICA method. Substituting the separation matrix into the following model:

[0081] Y = WX,

[0082] Where Y is the signal that is close to the tag signal after separation of M mixed signals, and W is an M×N dimensional instantaneous linear separation matrix.

[0083] In some embodiments of this application, the system further includes a verification module, which is used for:

[0084] If the signal separated by the first or second preset separation method does not contain any unidentified tags, then vehicle identification is performed based on the separated signal.

[0085] If the signal separated by the first or second preset separation method contains unidentified tags, then repeat steps one through four.

[0086] Those skilled in the art will understand that the modules in the system of the implementation scenario can be distributed throughout the system of the implementation scenario as described, or they can be modified to reside in one or more systems different from this implementation scenario. The modules of the above-mentioned implementation scenario can be merged into one module, or they can be further divided into multiple sub-modules.

[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for identifying electronic vehicle identification tags, characterized in that, Applied to a multi-antenna RFID system, where a preset number of antennas is defined, the method includes: Step 1: The reader sends a query command specifying the frame length. Tags within the reader's recognition range randomly select a time slot within the frame length range to respond to the reader's command and return an information packet. Step 2: If a collision time slot exists, after the previous frame ends, obtain the observation value in the previous frame, and obtain the number of tags based on the observation value in the previous frame and the preset tag estimation algorithm; Step 3: Based on the number of tags and the number of antennas, obtain the number of time slots for the next frame. Based on the number of time slots for the next frame, determine whether the number of antennas is greater than the number of tags in the time slots of the next frame. Step 4: If the number of antennas is greater than the number of tags in the next frame time slot, the first preset separation method is used to separate the tag signals, and vehicle identification is performed based on the separated signals. If the number of antennas is not greater than the number of tags in the next frame time slot, the second preset separation method is used to separate the tag signals, and vehicle identification is performed based on the separated signals. The collision time slot is the time slot of two or more tag return information packets; If the number of antennas is not greater than the number of tags in the next frame time slot, then the second preset separation method is used to separate the tag signals, specifically: Let there be N tags in a time slot, and the return signals of the N tags be S = [S1, S2, S3, ..., S2]. N The reader has M antennas, which receive M random mixed signals X = [X1, X2, X3, ..., X...] from N tags. M ], where X=AS, A is an N×M dimensional instantaneous linear mixture matrix with full column rank, and the separation matrix W is obtained by ICA method. Substituting the separation matrix into the following model: Y=WX, Where Y is the signal that is close to the tag signal after separation of M mixed signals, and W is an M×N dimensional instantaneous linear separation matrix.

2. The method as described in claim 1, characterized in that, The number of labels is obtained based on the observations in the previous frame and a preset label estimation algorithm, specifically as follows: The observed values ​​include the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of idle time slots is the number of time slots in which no tag returns a packet. The number of successful time slots is the number of time slots in which only one tag returns a packet. The number of collision time slots is the number of time slots in which two or more tags return packets. The probability of a collision slot appearing in the initial frame is calculated based on the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of tags is then roughly estimated based on the probability to obtain the first number of tags. If the first number of tags does not meet the preset requirements, a fine estimate is performed to obtain the number of tags. If the number of the first tag meets the preset requirement, then the number of the first tag will be used as the total number of tags.

3. The method as described in claim 1, characterized in that, The method further includes: If the signal separated by the first or second preset separation method does not contain any unidentified tags, then vehicle identification is performed based on the separated signal. If the signal separated by the first or second preset separation method contains unidentified tags, then repeat steps one through four.

4. A vehicle electronic identification system, characterized in that, Applied to a multi-antenna RFID system, the system has a preset number of antennas and includes: The response module is used by the reader to send a query command specifying the frame length. Tags within the reader's recognition range randomly select a time slot within the frame length range to respond to the reader's command and return an information packet. The estimation module is used to obtain the observation value in the previous frame after the previous frame ends if a collision time slot exists, and obtain the number of tags based on the observation value in the previous frame and a preset tag estimation algorithm. The judgment module is used to obtain the number of time slots in the next frame based on the number of tags and the number of antennas, and to determine whether the number of antennas is greater than the number of tags in the time slots of the next frame based on the number of time slots in the next frame. The separation module is used to separate the tag signal using a first preset separation method if the number of antennas is greater than the number of tags in the next frame time slot, and to identify the vehicle based on the separated signal; if the number of antennas is not greater than the number of tags in the next frame time slot, it is used to separate the tag signal using a second preset separation method, and to identify the vehicle based on the separated signal. The collision time slot is the time slot of two or more tag return information packets; The separation module is specifically used for: Let there be N tags in a time slot, and the return signals of the N tags be S = [S1, S2, S3, ..., S2]. N The reader has M antennas, which receive M random mixed signals X = [X1, X2, X3, ..., X...] from N tags. M ], where X=AS, A is an N×M dimensional instantaneous linear mixture matrix with full column rank, and the separation matrix W is obtained by ICA method. Substituting the separation matrix into the following model: Y=WX, Where Y is the signal that is close to the tag signal after separation of M mixed signals, and W is an M×N dimensional instantaneous linear separation matrix.

5. The system as described in claim 4, characterized in that, The estimation module is specifically used for: The observed values ​​include the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of idle time slots is the number of time slots in which no tag returns a packet. The number of successful time slots is the number of time slots in which only one tag returns a packet. The number of collision time slots is the number of time slots in which two or more tags return packets. The probability of a collision slot appearing in the initial frame is calculated based on the number of idle time slots, the number of successful time slots, and the number of collision time slots. The number of tags is then roughly estimated based on the probability to obtain the first number of tags. If the first number of tags does not meet the preset requirements, a fine estimate is performed to obtain the number of tags. If the number of the first tag meets the preset requirement, then the number of the first tag will be used as the total number of tags.

6. The system as described in claim 5, characterized in that, The system further includes a verification module, which is used for: If the signal separated by the first or second preset separation method does not contain any unidentified tags, then vehicle identification is performed based on the separated signal. If the signal separated by the first or second preset separation method contains unidentified tags, then repeat steps one through four.

Images