Label identification method and device based on filtering character string generation

Abstract

The invention discloses a label identification method and device based on filtering character string generation, and belongs to the technical field of wireless radio frequency identification. The method comprises the steps that a label list is set, a set of filtering character strings are generated through a filtering character string generation algorithm, and each filtering character string corresponds to a label in the label list; sending a Select command to the labels to select all the labels containing the filtering character strings in the EPC, and calculating the characteristic character strings of the labels; according to a selection label of the feature character string, performing label checking by sending a Query command, and determining that the label state is a stay label or a leave label; and selecting all tags which are matched with the filtered character string but do not contain the feature character string, sending a Query command to carry out standard checking operation, and collecting a new tag EPC. According to the invention, possible conflicts in the tag identification process can be reduced, and the identification efficiency and accuracy are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of radio frequency identification technology, and more specifically, to a tag recognition method and device based on filtered string generation. Background Art

[0002] In the application of radio frequency identification (RFID) technology, tag recognition is a crucial link, especially in a high-density and dynamically changing RFID environment. Currently, tag recognition methods based on RFID technology usually face various challenges such as tag quantity fluctuations, collision problems, and redundant data collection. A radio frequency identification system consists of a reader and multiple tags.

[0003] Existing tag recognition protocols mainly include direct tag ID collection protocols, tag hash-based recognition protocols, and bit pre-writing-based recognition protocols. However, these protocols have some deficiencies in commodity RFID systems, which limit their practical applications.

[0004] (1) Direct tag ID collection protocols. Such protocols are usually referred to as anti-collision methods, aiming to identify all tags within the detection range of the reader by directly retrieving the electronic product code (EPC) of the tags. Although these protocols effectively obtain the EPCs of known tags (i.e., staying tags and leaving tags) and store them in the tag list L, they cannot distinguish new tags from previously identified tags, especially between the staying tags and leaving tags maintained in the list L. Therefore, these protocols may lead to redundant EPC collection for the identified tags, thereby introducing inefficiency. This inefficiency problem is particularly obvious in a dynamic RFID environment, especially when the number of tags is constantly changing, resulting in a decline in system performance.

[0005] (2) Tag hash-based recognition protocols. To reduce redundant EPC collection, some protocols use hash functions to distinguish new tags from identified tags (staying tags and leaving tags) and process these two categories of tags separately. However, these protocols usually rely on complex on-tag hash functions that are beyond the capabilities of standard commercial off-the-shelf (COTS) tags. Although this method may be effective in some special enhanced tags (such as WISP tags), it is not applicable to standard COTS tags in most commodity RFID systems. Since standard COTS tags usually do not support complex hash operations, this makes tag hash-based recognition protocols unable to be widely applied in commodity RFID systems.

[0006] (3) Identification protocol based on bit pre-writing. In the prior art, identification methods based on pre-written data on COTS tags have received extensive attention. These methods generate hash values by utilizing the pre-written information on the tags (such as tag fingerprints or serial numbers) and allocate communication slots for the required tags, thus simplifying the identification process in commodity RFID systems. Protocols such as OPT-LR and MPS adopt this method to ensure that during EPC collection, communication slots are specifically allocated to new tags to avoid the collection of redundant data. However, these methods also have some problems: First, they generate a large pre-writing overhead, increasing the communication burden; Second, many COTS tags lack writable memory, resulting in these protocols may be incompatible with some tags; Finally, managing pre-written data increases the operational complexity of the system and brings additional communication overhead.

[0007] Existing tag identification protocols have certain limitations, which restrict their applicability in dynamic commodity RFID systems. Specifically, the main problems include: (1) Dependence on unsupported tag hash functions: Protocols based on tag hashing need to use complex hash functions, which are often infeasible for standard COTS tags, so these protocols are not applicable to most commodity RFID systems. (2) High pre-writing overhead: Protocols based on bit pre-writing need to write a large amount of pre-written data into the tags, which not only increases the communication overhead but may also lead to a decline in system performance. (3) Incompatibility with non-writable tags: Some protocols assume that tags have writable memory, but many COTS tags do not have this feature, which makes these protocols face compatibility problems in practical applications. Summary of the Invention

[0008] 1. Technical problems to be solved

[0009] In view of the deficiencies in the prior art, the present invention provides a tag identification method and device based on the generation of filter strings. Filter strings are generated based on the inherent randomness of the EPC and StoredCRC fields in RFID tags to achieve efficient tag identification in a dynamic environment.

[0010] 2. Technical solutions

[0011] The object of the present invention is achieved through the following technical solutions.

[0012] A tag identification method based on the generation of filter strings, comprising the following steps:

[0013] Set a tag list, and generate a set of filter strings through a filter string generation algorithm. Each filter string corresponds to a tag in the tag list;

[0014] Send a Select command to the tags to select all tags whose EPCs contain the filtering string, and calculate the feature strings of the tags;

[0015] Select tags according to the feature strings, and conduct a tag inventory by sending a Query command to confirm whether the tag status is a staying tag or a leaving tag;

[0016] Select all tags that match the filtering string but do not contain the feature string, send a Query command to conduct a tag inventory, and collect the EPCs of new tags.

[0017] As a further improvement of the present invention, for the tags in the tag list, perform the following steps:

[0018] Send a Select command to the tags to select all tags whose EPCs contain the filtering string;

[0019] Calculate the feature strings of the tags, including extracting substrings in the StoredCRC field of the tags as feature strings;

[0020] Send a Select command to the tags again to select the tags whose StoredCRC fields contain the feature strings;

[0021] Send a Query command to all tags whose StoredCRC fields contain the feature strings to conduct a tag inventory and identify the tag types.

[0022] As a further improvement of the present invention, generate a set of filtering strings through a filtering string generation algorithm, and the steps include:

[0023] Select the substring length range: calculate the minimum substring length and the maximum substring length according to the number of tags in the tag list;

[0024] Generate filtering strings: extract substrings from the EPC and StoredCRC fields, and determine whether the substring is a unique substring; if so, use it as the filtering string of the tag; if not, continue to increase the substring length until a filtering string is found;

[0025] Spare filtering string: if a unique substring cannot be found within the selected length range, use the EPC as the spare filtering string.

[0026] As a further improvement of the present invention, the filtering string includes the starting position within the EPC field, the length of the filtering string, and the specific position.

[0027] As a further improvement of the present invention, for staying tags, perform the following steps:

[0028] Issue the first Select command to select all tags that match the filtering string;

[0029] Calculate where r i represents the length of the i-th string, m 1.5 represents 1.5 to the power of the number of remaining tags, w represents the number of new tags, represents the length of the starting filtering string, and extract a substring of length r from the StoredCRC field of the tag to form a feature string; i

[0030] Issue the second Select command to select all tags that match the filtering characters but do not contain the feature string;

[0031] Issue a Query command to conduct a tag inventory and collect the new tag EPCs that do not match the feature string.

[0032] As a further improvement of the present invention, issue m more Select commands to isolate the tags that do not match the m filtering strings, and collect the new tag EPCs during the final tag inventory.

[0033] As a further improvement of the present invention, maintain the tag list, including:

[0034] The tag list contains the EPCs of all recognized tags and their statuses;

[0035] If the RFID system detects a new tag entering the detection range, it will add the tag to the tag list;

[0036] If the RFID system detects a tag leaving, it will remove the tag from the tag list.

[0037] As a further improvement of the present invention, during the verification phase, for each tag in the tag list, after issuing the Select command, calculate the feature string of the tag, and compare the StoredCRC field of the tag by issuing the second Select command to select the tags that meet the conditions;

[0038] If the tag does not respond to RN16, mark the tag as a leaving tag;

[0039] If the tag responds to RN16, mark the tag as a remaining tag and continue with the verification process of the next tag.

[0040] As a further improvement of the present invention, during the collection phase, issue the first Select command to select all tags that match the filtering string;

[0041] Issue a second Select command to select all tags that match the filtering string but do not contain the feature string;

[0042] Issue a Query command to conduct a tag inventory and collect new tag EPCs that do not match the feature string;

[0043] Issue another m Select commands to isolate tags that do not match the m filtering strings and collect new tag EPCs during the final tag inventory.

[0044] A tag recognition device generated based on a filtering string, comprising:

[0045] A generation module that sets a tag list and generates a set of filtering strings through a filtering string generation algorithm, with each filtering string corresponding to a tag in the tag list;

[0046] A selection module that issues a Select command to the tags to select all tags whose EPCs contain the filtering string and calculates the feature string of the tags;

[0047] A verification module that selects tags based on the feature string and conducts a tag inventory by sending a Query command to confirm whether the tag status is a staying tag or a leaving tag;

[0048] A collection module that selects all tags that match the filtering string but do not contain the feature string, issues a Query command to conduct a tag inventory, and collects new tag EPCs.

[0049] 3. Beneficial effects

[0050] Compared with the prior art, the advantages of the present invention are as follows:

[0051] (1) A tag recognition method and device generated based on a filtering string according to the present invention generate compact filtering strings based on the inherent randomness of the EPC and StoredCRC fields in RFID tags, which can effectively solve the problems of tag collision and redundant data collection existing in the existing RFID system, and significantly improve the efficiency and accuracy of tag recognition.

[0052] (2) A tag recognition method and device generated based on a filtering string according to the present invention further propose to dynamically maintain a tag list to distinguish new tags from known tags (staying tags and leaving tags) in order to avoid redundant data collection, especially in a dynamic RFID environment where the number of tags is constantly changing. In this way, the RFID system can timely identify and eliminate known tags, avoid repeated collection of EPC information of recognized tags, and significantly improve resource utilization.

[0053] (3) A tag recognition method and device based on filtered string generation according to the present invention optimize the tag verification and collection process, reduce unnecessary operations and calculations. The RFID system dynamically filters tags through multiple rounds of Select commands and Query commands, ensuring that the EPCs of all new tags are obtained in the shortest time, while avoiding redundant recognition of known tags, not only reducing data redundancy problems, but also further improving the speed and efficiency of tag recognition. Description of the Drawings

[0054] Figure 1 It is a flowchart of the method of the embodiment of the present invention;

[0055] Figure 2 It is a schematic structural diagram of the system of the embodiment of the present invention. Detailed Embodiments

[0056] The present invention will be described in detail below in conjunction with the drawings in the specification and specific embodiments.

[0057] Embodiment 1

[0058] In view of the problems of redundant communication overhead of tag recognition protocols, poor hardware compatibility, and insufficient dynamic adaptability in the prior art, this embodiment provides a tag recognition method based on filtered string generation. As Figure 1 shown, it includes the following steps: setting a tag list, generating a set of filtered strings through a filtered string generation algorithm, and each filtered string corresponds to a tag in the tag list; sending a Select command to the tags to select the tags whose EPCs contain the filtered strings, and calculating the feature strings of the tags; selecting tags according to the feature strings, and performing tag inventory by sending a Query command to confirm whether the tag status is a staying tag or a departed tag; selecting all tags that match the filtered strings but do not contain the feature strings, sending a Query command to perform tag inventory, and collecting the EPCs of new tags.

[0059] Specifically in this embodiment, the tag types in the RFID system are determined, including staying tags (Staying Tag) and departed tags (Departed Tag). According to the current tags within the range of the reader in the RFID system, an accurate and real-time dynamic tag list L is set. It should be noted that in this embodiment, the tag list also needs to be maintained, specifically including: the tag list contains the EPCs and their statuses of all recognized tags; if the RFID system discovers a new tag entering the detection range, the tag will be added to the tag list; if the RFID system detects a tag leaving, the tag will be removed from the tag list. Thus, by maintaining a real-time updated tag list, recording all tags within the current recognition range, including staying tags and departed tags, the status of the tags can be dynamically updated.

[0060] Thus, in this embodiment, by using a dynamically maintained tag list to distinguish new tags (New Tag) from known tags (staying tags and departed tags), the RFID system can promptly identify and eliminate known tags, avoiding duplicate collection of EPC information of identified tags and significantly improving resource utilization.

[0061] In this embodiment, it is also necessary to further verify the internal randomness of the EPC by analyzing the empirical entropy, including: (1) performing entropy analysis on the EPC field to confirm the existence of a large amount of randomness in the EPC field. Specifically, the EPC field is divided into 24 sub-arrays, each sub-array contains 4 bits, and the empirical entropy of each sub-array is calculated using data from multiple tags. Through entropy value analysis, the entropy values of the last 4 sub-arrays are 3.92 bits, 3.90 bits, 3.94 bits, and 3.95 bits respectively, indicating that each sub-array contains more than 3.9 random bits, thus confirming the existence of a large amount of randomness in the EPC field. (2) Performing entropy analysis on the StoredCRC field to verify the high randomness of the StoredCRC field. Specifically, the StoredCRC field is divided into 16 sub-arrays, each sub-array consists of a single bit, and the empirical entropy of the StoredCRC field is calculated. It is found that the empirical entropy of all bits is greater than 0.994, indicating that the StoredCRC field contains more than 15.9 random bits, thus verifying the high randomness of this field.

[0062] In this embodiment, for each tag in the tag list, the following steps are performed: The reader issues a Select command to the tag to select all tags whose EPC (Electronic Product Code) contains a filtering string; calculate the characteristic string of the tag. Specifically, a substring in the StoredCRC field of the tag is extracted as the characteristic string; the reader issues a Select command to the tag again to select all tags whose StoredCRC field contains the characteristic string; the reader issues a Query command to all tags whose StoredCRC field contains the characteristic string to perform a standard tag inventory operation to identify the tag type, that is, when the selected tag responds, it is confirmed as a "StayingTag", otherwise it is marked as a "Departed Tag".

[0063] In this embodiment, for staying tags, the following steps are performed: The reader issues the first Select command to select all tags that match the filtering string; calculate where r i represents the length of the i-th string, m 1.5 represents 1.5 to the power of the number of staying tags, and w represents the number of new tags. Indicates the starting filter string length and extracts a substring of length r from the StoredCRC field of the tag to form a feature string; the reader issues a second Select command to select all tags that match the filter characters but do not contain the feature string; the reader issues a Query command to inventory the tags and collect the new tag EPCs that do not match the feature string. The reader issues m additional Select commands to isolate the tags that do not match the m filter strings and collect the new tag EPCs during the final tag inventory. i It should be noted that traditional RFID identification methods usually directly compare the tag EPCs. However, when the number of tags is large, this method is prone to conflicts, which in turn affects the identification efficiency. In this embodiment, by utilizing the inherent randomness of the EPC field and the StoredCRC field in the RFID tag, a filtering string generation algorithm (NewFilter Generator, NFG) is used to dynamically generate filtering strings that meet the requirements, thereby effectively reducing tag conflicts.

[0064] In this embodiment, a set of filtering strings is generated through the filtering string generation algorithm, and each filtering string corresponds to a tag in the tag list. In this embodiment, the filtering string generation algorithm is used to generate filtering strings based on the EPC field and the StoredCRC field of the tag. This algorithm generates a unique and compact filtering string by selecting a substring of an appropriate length (i.e., random bits of the EPC and CRC).

[0065] The process of generating a set of filtering strings through the filtering string generation algorithm is as follows: Determine the minimum and maximum substring lengths for differentiating tags; for each tag and each possible substring length and position, check whether the substring is uniquely different from other tags; if a unique substring is found, generate a filtering string, including the starting position, length, and mask; if no unique substring is found, use the complete EPC of the tag as the filtering string; repeat this process until all tags are processed.

[0066]

[0067] ​Specifically, (1) Select the substring length range: Calculate the minimum substring length and the maximum substring length according to the number of tags in the tag list. This calculation method ensures that the generated filtering string can guarantee randomness and avoid generating overly long invalid substrings. (2) Generate the filtering string: Extract substrings from the EPC and StoredCRC fields and determine whether the substring is a unique substring; if so, use it as the filtering string for the tag; if not, continue to increase the substring length until a filtering string is found. (3) Backup filtering string: If a unique substring cannot be found within the selected length range, use the EPC as the backup filtering string. This step ensures that the tag can be correctly identified even in extreme cases.

[0068] In this embodiment, the filtering string includes the starting position within the EPC field, the length of the filtering string, and the specific positions involved. The process of obtaining the sampling statistical results for all intervals is as follows:

[0069] Define the minimum substring length L min and the maximum substring length L max , where L min = [log2(n ln(n))], L max = [log2(n ln(n))], and use the inherent randomness of the 16-bit EPC string to generate a compact filtering string for each tag in the tag list; check whether the given substring is unique by ensuring that no other tag t k ∈ tag list L shares this substring for uniquely identifying the tag; if the substring is not unique, the algorithm tests other substrings of the same length or increases the substring length; if the substring is unique, select it as the filtering string and continue to process the next tag; if no unique substring is found after calculating all substrings with lengths not exceeding the maximum substring length L max , select the complete 96-bit EPC as the filtering string.

[0070] Thus, in this embodiment, a compact filtering string is generated through the filtering string generation algorithm, which can effectively reduce the conflicts that may occur during the tag recognition process and significantly improve the efficiency and accuracy of tag recognition.

[0071] In this embodiment, during the verification phase, the RFID system sends a Select command to the tag to select the tags that match the filtering string. For the tags that meet the conditions, the RFID system calculates the feature string and further filters out the tags that do not meet the conditions. Through multiple rounds of Select commands and Query commands, the system gradually confirms the status of the tags (staying tags or departing tags). Specifically, for each tag in the tag list, after the reader sends a Select command, the feature string of the tag is calculated, and by sending a second Select command to compare the StoredCRC field of the tag, the tags that meet the conditions are selected; if the tag does not respond with RN16, the tag is marked as a Departed Tag; if the tag responds with RN16, the tag is marked as a Staying Tag, and the verification process for the next tag continues.

[0072] During the verification phase, the verification phase of the CETI protocol aims to effectively confirm the existence of each tag in the tag list, which is achieved through a selective reading process. The reader first uses a filtering string generation algorithm to generate a set F of n filtering strings, and each filtering string corresponds to a tag in the tag list. For each tag t i ∈ tag list L, where i ∈ {1, 2, …, n}, the reader performs the following steps:

[0073] (1) The reader sends the first Select command to select all tags whose EPC field contains the filtering string. In this embodiment, the command parameters are: Target = 1002, Action = 0, MemBank = 1, In this embodiment, since the StoredCRC field occupies the first 16 bits of the bank-01 memory, and the EPC starts from the 16th bit in the same memory bank, the offset of the pointer is set to 16, and the correct alignment of filtering tags using the EPC field is ensured through the offset;

[0074] (2) Calculate Select the index I ∈ [0, 16 - r i +1], extract a substring from the StoredCRC field to form the feature string Specifically, It jointly specifies the bit string from the I-th bit to the (I + r i -1)-th bit in the StoredCRC field of the tag as the feature string;

[0075] (3) The reader sends the second Select command with storedcrc containing the i-bit feature string G tiThe tagged one is the target. In this embodiment, the command parameters are: Target = 1002, Action = 2, MemBank = 1,

[0076] (4) The reader starts a single-slot inventory cycle by issuing a Query command with Q = 0. At this time, only the tags selected in steps (1) and (3) will respond;

[0077] (5) If no tag responds with RN16 during the inventory period, the reader identifies it as a departing tag and deletes it from the tag list. Otherwise, if a response is received, the reader confirms it as a staying tag.

[0078] In this embodiment, during the collection phase, all tags are inventoried through multiple Select commands and Query commands to ensure that the new tag EPC can be accurately collected. For known tags (staying tags and departing tags), redundant data collection is avoided to further improve the recognition efficiency. Specifically, the reader issues the first Select command to select all tags that match the filtering string; the reader issues the second Select command to select all tags that match the filtering string but do not contain the feature string; the reader issues a Query command to conduct a tag inventory and collect the new tag EPCs that do not match the feature string; the reader issues m additional Select commands to isolate the tags that do not match the m filtering strings and collect the new tag EPCs during the final tag inventory.

[0079] The collection phase starts with differentiating new tags and staying tags using filtering strings and feature strings derived from staying tags, and then proceeds with subsequent selective reading processes to collect new tag EPCs. The specific process is as follows:

[0080] (1) The reader uses a filtering string generation algorithm to generate a set F consisting of m filtering strings, where each filtering string corresponds to one of the m staying tags in the tag list;

[0081] (2) For each staying tag t i ∈ tag list L, where i ∈ {1, 2, …, m}, the reader performs the following sub-steps:

[0082] 2.1. The reader issues the first Select command to match all tags associated with the filtering string of the staying tag. This step is the mirror image of step (1) in the verification phase;

[0083] 2.2. The reader calculates r i ≤ log2(m 1.5 w) ≤ F2 i, and extract an r-bit substring from the StoredCRC field of the tag to form the feature string of the tag; i The substring of r bits forms the feature string of the tag;

[0084] 2.3. The reader issues a second Select command for all new tags that match the filter string but do not contain the feature string. In this step, set Action to 3, indicating that the matching tags reverse their SL flags without affecting the non-matching tags.

[0085] 2.4. The reader issues a Query command to start the tag inventory operation and collect the EPCs of the new tags that meet the filter string but do not meet the feature string;

[0086] (3) The reader issues m additional Select commands to isolate the new tags that do not match any of the m filter strings. Each Select command uses the filter-string feature string to deselect the new tags that match the filter string. After m commands, all new tags that do not match any of the m filter strings will be included in the final inventory operation, during which the new tag EPCs will be collected.

[0087] In this embodiment, during the verification phase and the collection phase, for any new tag, the probability that it shares an l-bit filter string with a known tag is 2 -1 , and the probability of sharing an r-bit feature string is 2 -r . By selecting the combined length of the filter string and the feature string (i.e., adjusting l + r), the reader can distinguish it from all other tags including new tags.

[0088] Thus, in this embodiment, in optimizing the tag verification and collection process, unnecessary operations and calculations are reduced. The FRID system dynamically filters tags through multiple rounds of Select commands and Query commands to ensure that all new tag EPCs are obtained in the shortest time, while avoiding redundant identification of known tags, which can not only reduce the data redundancy problem, but also further improve the speed and efficiency of tag identification.

[0089] A tag recognition method based on the generation of filter strings provided by this embodiment solves the problems of efficiency and compatibility in tag recognition in a dynamic environment by exploiting the inherent randomness of FRID tags, and provides a highly reliable and low-cost solution for scenarios such as intelligent manufacturing and intelligent logistics, with strong practicality and wide applicability.

[0090] Embodiment 2

[0091] Perform a lower bound analysis on a tag recognition method based on filtered string generation proposed in this embodiment. Specifically, derive the lower bound of the communication time for any c1g2-compatible scheme, which includes: (1) separating the resident tags from the new tags; (2) collecting the EPCs of the new tags.

[0092] Lower bound of task (1): Separating the standby tags from the new tags. To distinguish between resident tags and new tags, the reader must transmit at least m select commands, each using a different filtered string to separate a specific resident tag from the new tags. Each filtered string must be long enough to distinguish all w + m tags within the reader's range. Assuming there are w + m different tags, the minimum bit length required for the filtered string is log2(w + m) to ensure that each filtered string uniquely separates one resident tag from the others. Thus, the lower bound of the time required for this separation task is:

[0093] Since the existing C1G2 standard requires that only Select commands can be used for target tag separation and each resident tag requires a unique filtered string, this lower bound of time remains unchanged.

[0094] Lower bound of task (2): Collecting the EPCs of the new tags. Specifically, after separating the standby tags from the new tags, the next task is to collect the EPCs of the w new tags. According to the C1G2 standard, each new tag must transmit its EPC in a different time slot during the inventory. Therefore, at least w communication slots are required to prevent conflicts between EPC transmissions. The lower bound of the communication time for this task is:

[0095] Combined lower bound. By summing the lower bounds in formulas (1) and (2), the overall lower bound of the communication time for any c1g2-compatible acquisition scheme can be derived as: Thus, the theoretical lower bound of the communication time can be obtained as max(e(1 + ln(w / m)), 3).

[0096] Embodiment 3

[0097] As Figure 2As shown in the figure, this embodiment also provides a tag recognition system based on filtered strings, including a generation module, a selection module, a verification module, and a collection module. The generation module is used to set a tag list and generate a set of filtered strings through a filtered string generation algorithm, with each filtered string corresponding to a tag in the tag list. The selection module sends a Select command to the tags to select all tags whose EPCs contain the filtered strings and calculates the feature strings of the tags. The verification module selects tags based on the feature strings and conducts a tag inventory by sending a Query command to confirm whether the tag status is a stationary tag or a departing tag. The collection module selects all tags that match the filtered strings but do not contain the feature strings, sends a Query command to conduct a tag inventory, and collects the EPCs of new tags. A tag recognition device based on filtered string generation provided in this embodiment can implement any of the methods of the tag recognition method based on filtered string generation, and the specific working process of a tag recognition device based on filtered string generation can refer to the corresponding process in the embodiment of the tag recognition method based on filtered string generation. The methods and devices provided in this embodiment can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for example, the division of a certain module is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed connections or communication connections with each other can be indirect couplings or communication connections through some interfaces, devices, or units, and can also be electrical, mechanical, or other forms of connections.

[0098] The present invention and its implementation manners are schematically described above. This description is not restrictive. Without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. What is shown in the drawings is only one of the implementation manners of the present invention, and the actual structure is not limited thereto. Any reference numeral in the claims should not limit the claimed rights. Therefore, if those of ordinary skill in the art are inspired by this and, without departing from the purpose of this creation, design similar structural manners and embodiments to this technical solution without creative efforts, they shall fall within the protection scope of the present invention. In addition, the term "including" does not exclude other elements or steps, and the term "a" before an element does not exclude including "multiple" such elements. The multiple elements stated in the product claims can also be implemented by one element through software or hardware. First, second, etc. are used to indicate names and do not indicate any specific order.

Claims

1. A tag recognition method based on filtering string generation, comprising the following steps: Set the tag list, and generate a set of filter strings through the filter string generation algorithm, each filter string corresponds to a tag in the tag list; Send a Select command to the tag to select all tags that contain the filter string in the EPC, and calculate the characteristic string of the tag; Select tags based on the feature string, and perform tag inventory by sending a Query command to confirm whether the tag status is a stay tag or a leave tag; Select all tags that match the filter string but do not contain the feature string, issue a Query command to perform tag inventory, and collect new tag EPCs.

2. A tag recognition method based on filtering string generation according to claim 1, characterized in that: For a tag in the Tags list, perform the following steps: Send a Select command to the tag to select all tags whose EPCs contain the filter string; Calculating a characteristic string of the tag, including extracting a substring in the StoredCRC field of the tag as the characteristic string; Send the Select command to the tag again and select the tag that contains the characteristic string in the StoredCRC field; Send a Query command to all tags that contain characteristic strings in the StoredCRC field to perform tag inventory and identify tag types.

3. A tag recognition method based on filtering string generation according to claim 2, characterized in that: Generate a set of filter strings by using a filter string generation algorithm, the steps include: Select substring length range: Calculate the minimum substring length and maximum substring length based on the number of tags in the tag list; Generate filter string: extract substring from EPC and StoredCRC fields, and determine whether the substring is unique; if so, use it as the filter string for the tag; if not, continue to increase the substring length until the filter string is found; Backup filter string: If a unique substring cannot be found within the selected length range, use EPC as a backup filter string.

4. The tag recognition method based on filtering string generation according to claim 3 is characterized in that: The filter string includes the starting position in the EPC field, the length of the filter string, and the specific position.

5. The tag recognition method based on filtering string generation according to claim 1 is characterized in that: For the Dwell tab, perform the following steps: Issue the first Select command to select all tags that match the filter string; Calculate r i =[log2(m 1.5 w)-F2 i ], where r i Indicates the length of the i-th string, m 1.5 represents the 1.5th power of the number of staying tags, w represents the number of new tags, F2 i Indicates the length of the starting filter string and extracts the length r from the StoredCRC field of the tag i Substring of to form a characteristic string; Issue a second Select command to select all tags that match the filter string but do not contain the feature string; Issue a Query command to perform a tag inventory and collect new tag EPCs that do not match the feature string.

6. A tag recognition method based on filtering string generation according to claim 5, characterized in that: Then m Select commands are issued to isolate the tags that do not match the m filter strings, and the new tag EPCs are collected during the final tag inventory.

7. The tag recognition method based on filtering string generation according to claim 1 is characterized in that: Maintain the tag list, including: The tag list contains the EPCs of all identified tags and their status; If the RFID system finds a new tag entering the detection range, it will add the tag to the tag list; If the RFID system detects that the tag has left, it will remove the tag from the tag list.

8. The tag identification method based on filtering string generation according to claim 1 is characterized in that: In the verification phase, for each tag in the tag list, after issuing the Select command, the tag's characteristic string is calculated, and the StoredCRC field of the tag is compared by issuing a second Select command to select the tag that meets the conditions; If the tag does not respond to RN16, the tag is marked as a leave tag; If the tag responds to RN16, the tag is marked as a stay tag and the verification process of the next tag is continued.

9. The tag recognition method based on filtering string generation according to claim 1 is characterized in that: In the collection phase, the first Select command is issued to select all tags that match the filter string; Issue a second Select command to select all tags that match the filter string but do not contain the feature string; Issue a Query command to perform tag inventory and collect new tag EPCs that do not match the feature string; Then m Select commands are issued to isolate the tags that do not match the m filter strings, and the new tag EPCs are collected during the final tag inventory.

10. A tag recognition device based on filtering string generation, characterized in that: include: A generation module sets a tag list and generates a set of filter strings through a filter string generation algorithm, where each filter string corresponds to a tag in the tag list; The selection module sends a Select command to the tag to select all tags that contain the filter string in the EPC and calculate the characteristic string of the tag; The verification module selects tags according to the characteristic character string, and performs tag inventory by sending a Query command to confirm whether the tag status is a stay tag or a leave tag; The collection module selects all tags that match the filter string but do not contain the feature string, issues a Query command to perform a tag inventory, and collects new tag EPCs.

Images