Multi-classification label information collection method and system based on variable-length coding

Abstract

The invention discloses a multi-classification label information collection method and system based on variable-length coding. The method comprises the following steps: establishing a multi-classification system model consisting of commercial RFID readers and tags; defining a multi-classification label information collection problem under the model; the multi-classification information collection method based on the commercial RFID equipment specifically comprises the steps of designing a variable-length classification coding scheme which is efficient in storage and associated with label classification, designing masks according to target classification, and rapidly selecting labels of the target classification through a Select command compatible with the commercial RFID equipment to achieve efficient information collection. According to the invention, the problem that the internal memory of a commercial RFID passive tag is limited is solved by collecting tag information in a classified manner. The designed variable-length classification code is suitable for a multi-classification large-scale RFID system, the multi-classification label information collection efficiency is improved, and classification and management of article information are further optimized.

Description

Technical Field

[0001] The present invention relates to the field of target recognition and item information collection, and particularly to a method and system for collecting label information corresponding to target classification by using commercial devices in a multi-class commercial RFID system. Background Art

[0002] RFID (Radio Frequency Identification) technology is a non-contact automatic identification technology. It uses radio frequency signals and their spatial coupling transmission characteristics to realize the information identification of target objects attached with RFID tags. Compared with traditional barcodes or other identification technologies, RFID technology has significant advantages such as non-contact automatic identification, simultaneous reading of multiple tags, batch identification, and high-speed reading. These advantages have enabled RFID technology to be widely used in many fields, including but not limited to warehouse inventory, human-computer interaction, object positioning and tracking, and warehouse management. The popularization of RFID technology not only improves the management efficiency of various industries but also provides strong support for the development of intelligence and automation.

[0003] In a warehouse environment where RFID technology is applied, items are usually classified into different categories according to their attributes (such as item type, brand, batch, etc.), thus forming a multi-class RFID system. In a multi-class RFID system, a single item can belong to multiple categories simultaneously. For example, a MacBook can be classified as a laptop by type, as an Apple product by brand, and as an electronic product by attribute. This multi-classification method greatly optimizes the management of target objects and can efficiently execute tasks for specific categories, such as label loss detection, cardinality estimation, or sensor data collection. This meets the needs of practical applications: managers usually pay more attention to the information of specific categories of items rather than all items. For example, by querying a specific category, managers can quickly obtain the information of Apple MacBook Pro laptops produced in 2024 and later, which is much more efficient than querying all tags one by one.

[0004] In a multi - classification RFID system, to achieve effective item management based on classification, the key lies in efficiently collecting and processing information according to the classification of tags. In recent years, although some advanced methods have been proposed for the information collection problem in multi - classification RFID systems, these methods still have significant limitations in practical applications, especially in commodity RFID systems where they are difficult to directly implement. Specifically, the existing solutions mainly face the following limitations: (1) Incompatible with the EPC global Class1Gen2 standard and unable to be directly deployed in commercial RFID systems; (2) Only considering the simplified scenario where a tag belongs to a single classification and ignoring the actual need for precise management of items belonging to multiple classifications. Summary of the Invention

[0005] Object of the Invention: The present invention aims to address the deficiencies of the existing technology and proposes a method and system for collecting tag information for a multi - classification commercial RFID system, also known as a method and system for collecting multi - classification tag information based on variable - length coding. Based on real - world commercial RFID devices, this technology realizes the precise selection of tag information belonging to specific classifications, thereby achieving the classification management of tags and significantly improving the efficiency and accuracy of item management and operation.

[0006] To achieve the above - mentioned object of the invention, a method for collecting multi - classification tag information based on variable - length coding disclosed by the present invention includes the following steps:

[0007] Establish a multi - classification RFID system model consisting of a single reader and a set of tags. Each tag in the system has a unique ID for identifying the item to which it is attached. The tag set is represented as T = {t 1 ,t 2 ,…,t n}, where n is the total number of tags; the classification set is represented as C = {c 1 ,c 2 ,…,c m}, where m is the total number of classifications; each tag t i ∈T belongs to a classification subset c i , where c i ∈C; each classification c j ∈C contains a tag subset T j , where T j ∈T;

[0008] According to the established multi - classification RFID system model, define the multi - classification tag information collection problem as follows: Given a set of tags T, a set of classifications C, the maximum number of classifications m a associated with each tag, and the target query classification set C t , where How to encode and write the classification information associated with each tag, and collect the tags whose classifications in T match the elements in C t in C

[0009] Collect questions according to the defined multi-classification tag information, design a variable-length coding scheme for the affiliated classification of the associated tags, and only encode the affiliated classification information of each tag;

[0010] Design a mask for the target classification according to the coding scheme, generate the corresponding mask string based on the target query classification set, and the mask conforms to the tags of the target query classification;

[0011] Design a Select command according to the generated mask, select the set of tags that conform to the target query classification, and identify these tags, so as to realize the collection of multi-classification tag information.

[0012] Furthermore, design a variable-length coding scheme for the affiliated classification of the associated tags, and only encode the affiliated classification information of each tag, including:

[0013] The number of classifications in the system is m. For the i-th classification c i , 1 ≤ i ≤ m, encode it as the binary string corresponding to the decimal i, and the length of each binary string is bits;

[0014] The maximum number of classifications to which a tag belongs is m a , and each tag stores at most m a encoded affiliated classifications, and store these encoded information in ascending order of the classification ID. Before storing the classification information, all bits in the tag memory are initialized to 0, and the actual encoding length of the classification information of each tag changes according to the number of affiliated classifications and is not fixed.

[0015] Furthermore, design a mask for the target classification according to the coding scheme, and generate the corresponding mask string based on the target query classification set, including:

[0016] Define the target query classification set as C t ={C 1 , C 2 , …, C x}, where x is the number of elements in C t , and each element C i is called a sub-target classification set, which contains single classifications. 1 ≤ i ≤ x, C i represents the associated classification of the queried target tag, and the target tag contains all classifications in C i ;

[0017] Define the continuous classification segment as CCS. CCS is a subset composed of classifications with consecutive classification IDs, where the label must belong to all classifications within this classification segment. Formally, given a set of classifications C = {c x′ c i …c x c y …c j c y′}, i < j, when x + 1 = y (i + 1 ≤ x < y ≤ j - 1) and c x′ to c j and c i to c y′ are both not consecutive, the classifications c i to c j constitute a continuous classification segment with a length of j - i + 1;

[0018] Define the continuous classification range as CCR. CCR is a subset of classifications with consecutive classification IDs, where the label can belong to any one or more of these classifications. Formally, given a set of classifications C = {c x′ ,c i ,…,c x ,c y ,…,c j ,c y′}, i < j, when x + 1 = y (i + 1 ≤ x < y ≤ j - 1) and c x′ to c j and c i to c y′ are both not continuous classification ranges, the elements c i to c j constitute a CCR with a range of [i, j];

[0019] For each C t in C i , design masks respectively based on single-classification elements and multi-classification elements to generate a mask set for each C i ; the union of the individual mask sets for each C i constitutes the overall mask set of C t .

[0020] Furthermore, for each C t in C i , design masks respectively based on single-classification elements and multi-classification elements to generate a mask set for each C i , including:

[0021] For the single-classification element C i = {c i}, the mask is the binary string ID(c i ) corresponding to the classification ID;

[0022] For the case of co - selection of classifications in a multi - classification element, i.e., C i = c i …c j , the mask set consists of the union of the masks of CCS and the masks of non - CCS, where the mask of CCS is the concatenated string of the binary strings of its classification IDs, and the mask of non - CCS is the binary string of its classification ID;

[0023] For the case of individual selection of classifications in a multi - classification element, i.e., C i = {c i ,…,c j}, when these classification elements are not consecutive, each classification is selected one by one, and the mask design is the same as that of a single - classification element; when these classification elements are consecutive, the mask design method for CCR is adopted.

[0024] Furthermore, the mask design method for CCR includes:

[0025] For a CCR {c i ,…,c j}, when the minimum classification ID i > m a , the encoding of each element in the CCR may be stored at any position from 1 to m a in the tag memory, and the range query RQ(i, j) is used to generate a mask for the range i to j to select the tags containing the classifications in the CCR;

[0026] When the minimum classification ID i ≤ m a , the encoding of classification c i is stored at the positions from 1 to i in the tag memory. When j ≥ m a , for positions k = 1, 2,…, i, all classifications in the CCR may be stored, and the corresponding mask is RQ(i, j); for positions k = i + 1, i + 2,…, m a , elements in the CCR with classification IDs less than k cannot be stored, and the corresponding mask is RQ(k, j); when j < m a , the mask corresponding to the positions from 1 to j in the tag memory is RQ(i, j).

[0027] Furthermore, according to the generated mask, the Select command is designed, including the following steps:

[0028] For each element C t in C i , the Select command is based on C iWhether it is designed as a single - classification or multi - classification element, according to the RFID ultra - high - frequency international standard C1G2, use Session 2 of the Query command to query tags, that is, the first target field of the Select command is set to 2, and the tag query status is A or B;

[0029] In C i In the case of a single - classification element, when i > m a If a certain tag belongs to c i then the encoding of c i may be stored in its memory at any position from 1 to m a The first Select command is Flag←AB:S(2,a = 0,3,p 1 , where a = 0 means that the tag status that conforms to the mask is A, and other tags are B, and p 1 is the starting bit of the first position in the tag memory; the k - th Select command is Flag←A - :S(2,a = 1,3,p k , 2≤k≤m a where a = 1 means that the tag that conforms to the mask sets its status to A, and the tag that does not conform to the mask keeps its original status unchanged; p k is the starting bit of the k - th position in the tag memory; when i≤m a c i is stored in the first i positions of the tag memory, and i Select commands are required. The 1st and k - th Select commands are the same as above, 2≤k≤i;

[0030] In C i In the case of a multi - classification element, there are two scenarios for the design of the Select command: select tags that meet all classifications in C i or independently select tags that meet at least one classification. For the scenario of selecting tags that meet all classifications in C i the elements in C i are divided into CCS and non - CCS classifications. Among them, the Select command for non - CCS classifications is the same as in the single - classification case. For a CCS {c i ,…,c j}, when the maximum classification ID c j >m a the k - th Select command is where a = 2 means that the tag status that conforms to the mask remains unchanged, and the tag status that does not conform to the mask is B; p k is the starting bit of the k - th position; mask(c i-j ) is CCS {c i ,…,c jThe binary string of}; when j ≤ m a When, the value range of k is from 1 to i; otherwise, the value range of k is from 1 to m a -(j - i); if the Select command in C i is the first and k = 1, the action is AB, otherwise it is -B;

[0031] For the multi-classification elements where the label needs to satisfy at least one classification, the Select command is designed as follows: For the Select command design of non-CCR classification, it is the same as the single-classification case; for a CCR containing two or more classifications, the number of required Select commands is |M CCR |, where M CCR is the mask set of the CCR; for the M containing CCR the k-th element M CCR [k] in, 1 ≤ k ≤ |M CCR |, the Select command corresponding to the j-th mask in M CCR [k] is where p k is the starting bit at the k-th position, and M CCR [k][j] is the j-th mask in M CCR [k].

[0032] Furthermore, select the set of labels that meet the target query classification and identify these labels, including:

[0033] The reader sends the Select command mask label according to the specific requirements of C i , sets the label status that meets the mask to A, and the label status that does not meet the mask to B; then the reader sends the Query command to query these labels with status A and collects the information of the labels belonging to the target classification.

[0034] The present invention also provides a multi-classification label information collection system based on variable-length coding, including: an RFID reader, a plurality of RFID tags, and a processing device, and the processing device is configured to execute the steps of the above-mentioned multi-classification label information collection method based on variable-length coding.

[0035] The present invention also provides a computer device, including: one or more processors; a memory; and one or more programs, where the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and when the program is executed by the processor, it implements the steps of the above-mentioned multi-classification label information collection method based on variable-length coding.

[0036] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the multi-class label information collection method based on variable-length coding as described above are implemented.

[0037] Beneficial effects: (1) Focusing on the information collection requirements of multi-class RFID systems in practical applications and on the key issue of label information collection in multi-class commercial RFID systems, the present invention innovatively proposes an efficient label information collection method based on existing commercial RFID devices. Compared with existing information collection methods, the greatest feature of the present invention is that it takes into account that in practical applications, labels belong to multiple categories, and accordingly designs a multi-class label information method compatible with existing commercial RFID devices. By designing a variable-length information coding scheme for storing label membership categories efficiently and designing a Select command compatible with the C1G2 standard, this scheme can quickly screen and identify labels belonging to the target category from a large number of labels, effectively avoiding the communication burden caused by querying all labels, thus realizing efficient and accurate multi-class label information collection and significantly reducing economic losses caused by classification confusion or misjudgment. (2) The present invention further creates a prototype of a multi-class commercial RFID system, implements the proposed multi-class label information collection strategy based on commercial RFID devices, avoids interference from labels of non-target categories, significantly improves the information collection efficiency, and reduces the possible risks in the operation of the multi-class commercial RFID system. Description of the Drawings

[0038] Figure 1 is a prototype system for label information collection for commercial multi-class RFID systems.

[0039] Figure 2 is a flowchart of the multi-class label information collection method based on variable-length coding.

[0040] Figure 3 is a schematic diagram of mask design based on variable-length coding. Detailed Embodiments

[0041] The technical solution of the present invention will be further described below with reference to the drawings.

[0042] The present invention constructs an experimental platform for multi-class label information collection implemented based on commercial RFID devices, as Figure 1As shown in the figure, on the left side of the figure are the specifications of the RFID devices used in the experiment. The reader model is Alien F800, and the tag models are passive tags Alien 9962 and Alien 9940 with Higgs9 chips. On the right side is the built experimental platform. The tags are used in the system to identify the items they are attached to. The reader is connected to a directional antenna with a gain of 9dBic, model Laird S9028, and operating frequency of 920MHz. The reader communicates with the tags through the antenna to read the tag data. The computer control terminal is responsible for receiving the data collected by the reader and analyzing and processing the data. Based on this experimental platform, the present invention provides a multi-class information collection method based on variable-length coding. This method is mainly executed by the computer for the specific method process and controls the communication process between the reader and the tags to achieve the collection of tag information, such as Figure 2 as shown, including the following content.

[0043] Step 1, establish a multi-class RFID system model consisting of a single reader and a group of tags on the computer side. Each tag in the system has a unique ID to identify the item it is attached to. The tag set is represented as T = {t 1 , t 2 , …, t n}, where n is the total number of tags; the classification set is represented as C = {c 1 , c 2 , …, c m}, where m is the total number of classifications. Each tag t i ∈ T belongs to a classification subset c i , where c i ∈ C. Each classification c j ∈ C contains a tag subset T j , where T j ∈ T;

[0044] Step 2, according to the multi-class RFID system model created in Step 1, define the multi-class tag information collection problem as follows: Given a set of tags T, a set of classifications C, the maximum number of classifications m a associated with each tag, and the target query classification set C t , where how to effectively encode and write the classification information associated with each tag, and efficiently collect the tags in T whose classifications match the query elements in C t ;

[0045] Step 3, according to the multi-class tag information collection problem defined in Step 2, design a variable-length coding scheme for associating tag membership classifications, and only encode the membership classification information of each tag;

[0046] According to the implementation manner of the present invention, Step 3 includes the following steps:

[0047] Step 3-1: The number of classifications in the system is m. For the i-th (1 ≤ i ≤ m) classification c i , it is encoded into a binary string corresponding to the decimal i, and the length of each binary string is bits;

[0048] Step 3-2: The maximum number of classifications to which a label belongs is m a , and each label stores at most m a encoded membership classifications, and stores this encoded information in ascending order of classification ID. Before storing the classification information, all bits in the label memory have an initial value of 0. The present invention only encodes the membership classifications of each label. The actual length of encoding the classification information of each label varies with the number of membership classifications and is not fixed. For example, if a label belongs to only one classification, its encoding length is the length of one classification encoding, that is, And for a label that belongs to 3 classifications, its encoding length is the length of 3 classification encodings, that is, For example, if there are 10 classifications in the system, the encoding length of each classification is Suppose the membership classifications of a label are 2 and 4, then the classification encoding of this label is 0010 0100; if the membership classifications of another label are 3, 6, 8, then its classification encoding is 00110110 1000. It can be seen that the encoding lengths of the two labels are not the same.

[0049] Step 4: Design a mask for the target classification according to the encoding scheme in Step 3, generate a corresponding mask string based on the target query classification set, and mask the labels that conform to the target query classification;

[0050] Step 4-1: Define the target query classification set as C t ={C 1 , C 2 , …, C x}, where x is the number of elements in C t , and each element C i (1 ≤ i ≤ x) is called a sub-target classification set, which contains single classifications. C i represents the associated classification of the target label being queried, and the target label contains all the classifications in C i .

[0051] Step 4-2: Define a Consecutive Category Segment (CCS). CCS is a subset composed of classifications with consecutive classification IDs, and a label must belong to all the classifications within this classification segment. Formally, given a set of classifications C = {c x′c i …c x c y …c j c y′}(i < j). When x + 1 = y (i + 1 ≤ x < y ≤ j - 1) and c x′ to c j and c i to c y′ are not consecutive, classify c i to c j to form a consecutive classification segment with a length of j - i + 1;

[0052] Step 4 - 3, define the Consecutive Category Range as CCR. CCR is a subset of categories with consecutive category IDs, where the labels can belong to any one or more of these categories. Formally, given a set of categories C = {c x′ , c i , …, c x , c y , …, c j , c y′}(i < j), when x + 1 = y (i + 1 ≤ x < y ≤ j - 1) and c x′ to c j and c i to c y′ are not consecutive category ranges, the elements c i to c j form a CCR with a range of [i, j];

[0053] Step 4 - 4, for each C t in C i , design masks based on single - category elements and multi - category elements respectively, and generate a mask set for each C i ;

[0054] Specifically include:

[0055] Step 4 - 4 - 1, for the single - category element C i = {c i}, the mask is the binary string ID corresponding to the category ID (ID(c i ));

[0056] Step 4 - 4 - 2, for the case of common selection of categories in multi - category elements, that is, C i = c i …c j , the mask set is composed of the union of the masks of CCS and the masks of non - CCS. The mask of CCS is the concatenated string of the binary strings of its category IDs, and the mask of non - CCS is the binary string of its category ID;

[0057] Step 4-4-3. For the case of individual selection of categories in multi-category elements, i.e., C i ={c i ,…,c j}, when these category elements are not continuous, each category is selected one by one, and the mask design is the same as that for single-category elements in Step 4-3-1;

[0058] Step 4-4-4. For the case of individual selection of categories in multi-category elements, i.e., C i ={c i ,…,c j}, when these category elements are continuous, an efficient mask design method for CCR is adopted;

[0059] The described efficient mask design method for CCR includes the following steps:

[0060] Step 4-4-4-1. For a CCR {c i ,…,c j}, when the minimum category IDi > m a , the encoding of each element in the CCR may be stored at any position from 1 to m a in the tag memory. A range query RQ(i,j) is used to generate a mask for the range i to j to select the tags containing the categories in the CCR;

[0061] Step 4-4-4-2. When the minimum category IDi ≤ m a , the encoding of category c i is stored at positions from 1 to i in the tag memory. When j ≥ m a , for positions k = 1, 2, …, i, all categories in the CCR may be stored, and the corresponding mask is RQ(i,j); for positions k = i + 1, i + 2, …, m a , elements with category IDs less than k in the CCR cannot be stored, and the corresponding mask is RQ(k,k). When k < m a , the mask corresponding to positions 1 to k in the tag memory is RQ(i,j).

[0062] Step 4-5. The union of the individual mask sets of each C i constitutes the overall mask set of C t .

[0063] Next, taking Figure 3 as an example, the mask design process is further illustrated. Figure 3 's system consists of 13 categories c 1-13 and 5 tags t 1-5 , and each tag contains at most 4 categories, i.e., m a= 4. The target query classification set is C t = {{c 2 , c 3 , c 4}, {c 8 c 9}}, where the sub-target classification set {c 2 , c 3 , c 4} forms a CCR, and the corresponding mask is RQ(2, 4). RQ(2, 4) means finding the common prefix of the binary strings of 2, 3, and 4. For strings that have no common prefix with other strings, the mask is itself. Therefore, the masks generated by RQ(2, 4) are "001" and "0100" respectively. For the sub-target classification set {c 8 c 9}, c 8 c 9 forms a CCS, and the mask is the concatenation of the binary encodings of the two classifications, which is "10001001".

[0064] Step 5, design the Select command according to the mask generated in Step 4, select the set of tags that meet the target query classification, and further identify these tags, so as to realize the collection of multi-classification tag information.

[0065] For each element c t in c i (1 ≤ i ≤ x), the Select command is designed according to whether C i is a single-classification or multi-classification element. According to the RFID ultra-high frequency international standard C1G2, use Session 2 of the Query command to query the tags, that is, the first target field of the Select command is set to 2, and the tag query status is A or B;

[0066] First, consider the case where C i is a single-classification element. When i > m a , if a certain tag belongs to c i , then the encoding of c i may be stored at any position from 1 to m a in its memory. Therefore, m a Select commands are required to cover all positions from 1 to m a to select the tags containing c i . The first Select command is where a = 0 indicates that the tag status that meets the mask is A, and other tags are B, and p 1 is the starting bit of the first position in the tag memory. After the reader sends this Select command, starting from the p 1 bit, it is compared with mask(ci ) The matching tags set their query status to A, while the remaining tags set their status to B. For the kth (2 ≤ k ≤ m a ) Select command is where a = 1 means that the tags that match the mask set their status to A, and the tags that do not match the mask keep their original status unchanged; p k is the starting bit at the kth position in the tag memory. When i ≤ m a , c i is stored in the first i positions of the tag memory, and i Select commands are required. The 1st and kth (2 ≤ k ≤ i) Select commands are the same as above.

[0067] Then consider the case where C i is a multi-classification element. The Select command is designed with two scenarios: selecting the tags that satisfy all the classifications in C i or independently selecting the tags that satisfy at least one classification. For the scenario of selecting the tags that satisfy all the classifications in C i , the elements in C i are divided into CCS and non-CCS classifications. The Select commands for non-CCS classifications are the same as in the single-classification case. For a CCS {c i , …, c j}, when the maximum classification ID c j > m a , the kth Select command is where a = 2 means that the status of the tags that match the mask remains unchanged, and the status of the tags that do not match the mask is B; p k is the starting bit at the kth position; mask(c i-j ) is the binary string of CCS {c i , …, c j}. When j ≤ m a , the value range of k is from 1 to i; otherwise, the value range of k is from 1 to m a -(j - i). In addition, if the Select command in C i is the first and k = 1, the action is AB, and in other cases it is -B.

[0068] For multi-classification elements where the tag needs to satisfy at least one classification, the Select command is designed as follows. For the Select commands of non-CCR classifications, the design is the same as in the single-classification case. For a CCR with two or more classifications, the number of Select commands required is |M CCR |, where M CCR is the mask set of the CCR. For M containing CCR masks, for the kth (1 ≤ k ≤ |MCCR |) Element M CCR [k], M CCR [k] the j-th one (1 ≤ j ≤ ) The Select command corresponding to the mask is where p k is the starting bit at the k-th position, M CCR [k][j] is the j-th mask in M CCR [k].

[0069] In the Figure 3 example, for the first sub-goal query classification set {c 2 , c 3 , c 4}, the masks are "001" and "0100" respectively. The Select command S1 passes through the mask "001" and compares the first 3 bits of the label memory with the encoding of c 2 to select the label t 2 belonging to c 4 . Similarly, S3 selects the label t 2 for c 1 , and S5 also selects t 3 for c 1 because t 1 belongs to both the classification c 2 and c 3 . S2 uses the mask "0100" to select the label t 4 belonging to the classification c 5 . By comparing the number of Select commands with and without using CCR, it can be seen that CCR reduces the number of Select commands from 9 to 7, optimizing the masking process. For the second sub-goal query classification set {c 8 c 9}, the mask is "10001001", and the corresponding Select command uses this mask to select the label t 8 and c 9 belonging to both the classification c 3 .

[0070] The reader sends the Select command mask label according to the specific requirements of C i , sets the label status that conforms to the mask to A, and sets the label status that does not conform to the mask to B. Then it reads and sends the Query command to query these labels with status A, and collects the label information of the labels belonging to the target classification label.

[0071] The present invention further provides a multi-class label information collection system based on variable-length coding, including: an RFID reader, a plurality of RFID tags, and a processing device, where the processing device is configured to execute the steps of the multi-class label information collection method based on variable-length coding as described above.

[0072] The present invention further provides a computer device, including: one or more processors; a memory; and one or more programs, where the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and when the programs are executed by the processors, the steps of the multi-class label information collection method based on variable-length coding as described above are implemented.

[0073] The present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the multi-class label information collection method based on variable-length coding as described above are implemented.

[0074] The present invention provides a method and a system for collecting label information for a multi-class commercial RFID system. There are many methods and ways to specifically implement this technical solution. The above description is only a preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention. Each component not clearly defined in this embodiment can be implemented by the prior art.

Claims

1. A method for collecting multi-classification label information based on variable-length coding, characterized in that: The method comprises the following steps: A multi-classification RFID system model consisting of a single reader and a set of tags is established. Each tag in the system has a unique ID to identify the item it is attached to. The tag set is represented by T = {t1, t2, …, t n }, where n is the total number of labels; the classification set is represented by C = {c1, c2, ..., c m }, where m is the total number of categories; each label t i ∈T belongs to a classification subset c i , where c i ∈C; each category c j ∈C contains a label subset T j , where T j ∈T; According to the established multi-classification RFID system model, the multi-classification tag information collection problem is defined as follows: Given a set of tags T, a set of categories C, and the maximum number of categories m associated with each tag a And the target query classification set C t ,in How to encode and write the classification information associated with each label and collect the classification information in T and C t Query the tag information that matches the element; According to the defined multi-classification label information collection problem, a variable-length encoding scheme for associating labels to belong to categories is designed, and only the category information of each label is encoded; Design a target classification mask according to the encoding scheme, generate a corresponding mask string based on the target query classification set, and the mask conforms to the label of the target query classification; The Select command is designed based on the generated mask to select the set of labels that meet the target query classification and identify these labels, thereby realizing the collection of multi-classification label information.

2. The method according to claim 1, characterized in that Design a variable-length coding scheme for associating labels to classify, and only encode the classification information of each label, including: The number of categories in the system is m, and for the i-th category c i , 1≤i≤m, encode it into a binary string corresponding to decimal i, the length of each binary string is Bit; The maximum number of categories a label belongs to is m a , each tag can store up to m a The tags are coded in ascending order of the category ID. Before storing the category information, all tags in the tag memory are The initial value of the bit is 0. The actual classification information encoding length of each tag varies according to the number of categories it belongs to and is not fixed.

3. The method according to claim 1, characterized in that Design the target classification mask according to the encoding scheme, and generate the corresponding mask string based on the target query classification set, including: Define the target query classification set as C t ={C1,C2,…,C x }, where x is C t The number of elements in C i It is called the sub-target classification set, which contains single classification, 1≤i≤x, C i Indicates the associated classification of the queried target label, the target label contains C i All categories in; Define a continuous classification segment as CCS. A CCS is a subset composed of classifications with consecutive classification IDs, where the labels must belong to all classifications within this classification segment. Formally, given a set of classifications C = {c x′ c i …c x c y …c j c y′}, i < j. When x + 1 = y (i + 1 ≤ x < y ≤ j - 1) and c x′ to c j and c i to c y′ are both not consecutive, the classifications c i to c j constitute a continuous classification segment with a length of j - i + 1; Define the continuous classification range as CCR. CCR is a subset of classifications with consecutive classification IDs, where a label can belong to any one or more of these classifications; formally, given a set of classifications C = {c x′ , c i , …, c x , c y , …, c j , c y′}, i < j, when x + 1 = y (i + 1 ≤ x < y ≤ j - 1) and c x′ to c j and c i to c y′ are not continuous classification ranges, the elements c i to c j form a CCR with a range of [i, j]; For C t Each C i , respectively design masks based on single-classification elements and multi-classification elements to generate each C i The mask set of each C i The union of the individual mask sets of t The overall mask set.

4. The method according to claim 3, characterized in that For C t Each C i , respectively design masks based on single-classification elements and multi-classification elements to generate each C i The mask set includes: For a single-category element C i ={c i }, the mask is the binary string ID corresponding to the category ID (c i ); For the case of common selection of categories in multi-category elements, that is, C i =c i …c j , the mask set consists of the union of the mask of CCS and the mask of non-CCS, where the mask of CCS is the concatenated string of the binary string of its classification ID, and the mask of non-CCS is the binary string of its classification ID; For the case of single selection of categories in multi-category elements, that is, C i ={c i ,…,c j }, when these classification elements are discontinuous, each classification is selected one by one, and the mask design is the same as the case of a single classification element; when these classification elements are continuous, the mask design method for CCR is adopted.

5. The method according to claim 4, characterized in that The mask design method for CCR includes: For a CCR{c i ,…,c j }, when the minimum category IDi>m a When the encoding of each element in the CCR may be stored in the tag memory from 1 to m a At any position of the CCR, use the range query RQ(i,j) to generate a mask for the range i to j to select the label containing the classification in the CCR; When the minimum classification IDi≤m a When, classification c i The encoding is stored in the tag memory at positions from 1 to i, when j ≥ m a When, for position k = 1, 2, ..., i, all categories in CCR may be stored, and the corresponding mask is RQ(i, j); for position k = i+1, i+2, ..., m a , it is impossible to store elements with category ID less than k in CCR, and the corresponding mask is RQ(k,j); when j <m a When , the mask corresponding to the position from 1 to j in the tag memory is RQ(i,j).

6. The method according to claim 5, characterized in that Designing the Select command based on the generated mask includes the following steps: For C t Each element C in i The Select command is based on C i Whether it is designed for single-classification or multi-classification elements, according to the RFID UHF international standard C1G2, use Session 2 of the Query command to query the tag, that is, the first target field of the Select command is set to 2, and the tag query status is A or B; In C i In the case of a single-category element, when i>m a When a tag belongs to c i , then c i The encoding may be stored in its memory from 1 to m a Any position of the first Select command is Where a=0 means that the state of the tag that meets the mask is A, and the other tags are B. p1 is the starting bit of the first position in the tag memory; the kth Select command is Where a=1 means that the labels that meet the mask will be set to state A, and the labels that do not meet the mask will remain in their original state; p k is the starting bit of the kth position in the tag memory; when i≤m a When c i The first i positions stored in the tag memory require i Select commands, and the first and kth Select commands are the same as above 2≤k≤i; In C i In the case of multi-classification elements, the Select command is designed for two scenarios: select the elements that meet C i All the labels of the categories in or independently select the label that satisfies at least one category. For the selection that satisfies C i The scene with all the labels in C i The elements in are divided into CCS and non-CCS categories. The Select command for non-CCS categories is the same as that for single categories. For a CCS{c i ,…,c j }, when the maximum category IDc j >m a When , the kth Select command is Where a=2 means that the state of the label that meets the mask remains unchanged, and the state of the label that does not meet the mask is B; p k is the starting position of the kth position; mask(c i-j ) is CCS{c i ,…,c j } binary string; when j≤m a When , the value of k ranges from 1 to i; otherwise, the value of k ranges from 1 to m a -(ji); if C i If the Select command is the first one and k=1, the action is AB, otherwise it is -B; For multi-category elements whose labels need to satisfy at least one category, the Select command is designed as follows: The Select command design for non-CCR categories is the same as that for single categories; for a CCR containing two or more categories, the number of Select commands required is |M CCR |, where M CCR is the CCR mask set; for The mask M CCR The kth element M in CcR [k],1≤k≤|M CCR |, M CCR The Select command corresponding to the jth mask in [k] is where p k is the starting position of the kth position, M CCR [k][j] is M CCR The j-th mask in [k].

7. The method according to claim 1, characterized in that Select a set of tags that match the target query classification and identify these tags, including: By the reader according to C i The reader sends a Select command to mask the tags according to the specific needs, sets the status of the tags that meet the mask to A, and sets the status of the tags that do not meet the mask to B; then the reader sends a Query command to query the tags with status A and collect information belonging to the target classification tags.

8. A multi-classification label information collection system based on variable length coding, characterized in that: include: An RFID reader, a plurality of RFID tags and a processing device, wherein the processing device is configured to execute the steps of the variable-length coding-based multi-classification tag information collection method according to any one of claims 1 to 7.

9. A computer device, characterized in that: include: one or more processors; Memory; And one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and when the programs are executed by the processors, the steps of the multi-classification label information collection method based on variable-length coding as described in any one of claims 1-7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for collecting multi-classification label information based on variable-length coding as described in any one of claims 1 to 7 are implemented.

Images