Continuous tag identification method and system based on COTS RFID

Abstract

The invention provides a continuous tag identification method and system based on COTS RFID, and relates to the technical field of RFID. According to the method, a label identity management system is established by pre-distributing static identifiers to a label user storage area, and accurate screening and grouping of labels are realized by matching a standard selection instruction with specific mask parameters; an optimized filtering mechanism is adopted to distinguish known tags from unknown tags, and the tag state is confirmed and updated step by step through a staged polling strategy; wherein a static identification pre-storage mechanism obviously reduces performance requirements on label hardware, a selective polling method based on mask matching improves processing efficiency, a multi-stage fingerprint filtering structure ensures identification accuracy, and the design completely following a C1G2 protocol specification enables a system to be directly deployed in an existing RFID infrastructure.

Description

Technical Field

[0001] The present invention relates to the technical field of tag identification, and in particular to a continuous tag identification method and system based on COTS RFID. Background Art

[0002] With the rapid development of the Internet of Things (IoT) and smart manufacturing, RFID (Radio Frequency Identification) technology has become a crucial tool in logistics tracking, warehouse management, and supply chain optimization. By automatically identifying tag information through radio waves, RFID systems can identify multiple tags within range without human intervention and monitor item status in real time. For example, in warehouse management, RFID readers can automatically scan the tags of incoming and outgoing goods, enabling digital inventory management. In smart retail, RFID tags help merchants track product sales in real time, automatically adjust inventory, and optimize replenishment strategies. In manufacturing, RFID systems on production lines can efficiently track the status of products at each processing step, ensuring traceability of product quality.

[0003] However, due to the limited communication range between RFID tags and readers, tags frequently enter and exit the reader's range in dynamic environments such as logistics and production lines. This requires RFID systems to continuously run multiple recognition cycles to maintain an inventory of tags within range and quickly detect new or missing tags. For example, in warehouse management, due to the frequent handling of goods, some tags may temporarily leave the reader's reading area. Newly entered tags need to be promptly identified and added to the system inventory.

[0004] In practical applications, efficiently and cost-effectively identifying new and missing tags within limited communication time is a key challenge for mobile RFID systems. Existing technologies primarily improve recognition accuracy by increasing the reading frequency, but this approach often incurs additional communication and computational costs, making it difficult to meet the real-time requirements of highly dynamic environments. Many traditional protocols focus on a single task, such as identifying only new tags or detecting only missing tags, failing to simultaneously complete both tasks within the same communication cycle. This clearly limits their application in highly dynamic scenarios such as logistics and supply chain management.

[0005] Furthermore, the real-time requirements of mobile RFID systems make traditional periodic tag recognition methods difficult to meet the needs of dynamic environments. In practical application scenarios, RFID tags may not be successfully read within a certain recognition cycle due to occlusion, interference, or signal attenuation, thus affecting the accuracy and stability of the system. Existing research mainly uses redundant reading or multi-polling mechanisms to improve the recognition success rate, but these methods often significantly increase the communication burden and computational cost of the system, reducing overall efficiency. In fields such as logistics, warehousing, and supply chain management, the system needs to process large amounts of tag data in a short period of time and ensure the accuracy of the recognition results. Therefore, an efficient and low-overhead recognition method is urgently needed. Summary of the Invention

[0006] 1. Technical problem to be solved by the invention

[0007] This paper provides a continuous tag identification method and system based on COTS RFID. By preloading static order numbers and optimizing standard command sequences, this paper designs a tag identification mechanism compatible with commercial RFID devices. It also proposes a C1G2-compatible protocol, enabling efficient identification and management of tag states in dynamic environments. This tag identification method retains the theoretical advantages of existing advanced protocols while perfectly adapting to existing C1G2-standard equipment, providing important technical support for the industrial application of RFID technology.

[0008] 2. Technical solution

[0009] To achieve the above objectives, the present invention proposes the following technical solutions:

[0010] A continuous tag identification method based on COTS RFID of the present invention comprises the following steps:

[0011] Step S1: At the beginning of the reading cycle, the order number and fingerprint are stored in the C1G2 compatible version of the known tag and the unknown tag;

[0012] Step S2: Pre-store a unique static identifier for each known tag, selectively poll the known tags, detect missing tags within the current identification range, and build a fingerprint-based filtering structure to separate and identify unknown tags; based on the polling and filtering results, further confirm the remaining tags;

[0013] Step S3: Calculate the theoretical lower limit of the communication time using the communication time calculation model, and guide the optimization of protocol parameters based on the theoretical lower limit value to improve the tag reading order and filtering mechanism;

[0014] Step S4: The reader generates a tag status report based on the data results, provides real-time monitoring data to the user, and supports dynamic adjustment of the RFID system.

[0015] Furthermore, the known tag is a tag that has been registered in the system database and assigned a unique static identifier, the unknown tag is a new tag that has not been registered in the system or assigned a static identifier; the missing tag is a tag that has been registered in the system database but has not responded within the identification period.

[0016] Furthermore, step S2 includes three operation stages:

[0017] In the first phase, the reader screens known tags through a static identifier-driven masked polling mechanism to identify missing tags and avoid redundant communications caused by unresponsive tags;

[0018] In the second stage, the reader builds a fingerprint filtering structure based on the static identifier index, and guides the tag to generate a local fingerprint by broadcasting a random factor, thereby filtering out unknown tags.

[0019] In the third phase, the reader again performs verification polling on the remaining tags in a one-to-one manner to avoid the waste of collision time slots generated in the traditional frame structure.

[0020] Furthermore, the following steps are performed in the first phase:

[0021] a. The reader R maintains a tag list L, assigns a unique static identifier to each known tag, and writes it to the user storage area;

[0022] b. Each known tag t sends a predefined response signal "1" in the corresponding time slot according to the stored static identifier;

[0023] c. The reader R monitors the time slot response in real time. If a "1" signal is not received in a time slot, the tag corresponding to the time slot is marked as "missing";

[0024] d. The reader R constructs an n-bit response array S based on the detection results, identifies the response status of each tag, and broadcasts it to all tags;

[0025] e. Each tag updates its local state variable according to the response array S and records whether its own response status is normal.

[0026] Furthermore, in the second phase, the following steps are performed:

[0027] a. The reader R generates and broadcasts a random number v and fingerprint length parameter D as input for the tag to calculate the identity fingerprint;

[0028] b. After each tag t receives the random number v, it calculates the fingerprint value h(t) based on its own ID and the random number v;

[0029] c. The reader R constructs a filtering data structure F containing m units for the fingerprints h(t) of all known tags, where F[i] stores the fingerprint of the tag corresponding to the i-th static identifier;

[0030] d. Each tag t searches for its corresponding position in F according to its local static identifier. If the comparison result is consistent with the local fingerprint, it is marked as a "known tag", otherwise it is marked as an "unknown tag".

[0031] Furthermore, in the third phase the following steps are performed:

[0032] a. The reader R restarts a frame consisting of k time slots, where k is the number of known tags to be verified, and each time slot is synchronized with the static identifier of the tag;

[0033] b. Each tag t sends a "1" signal in the corresponding time slot according to its static identifier to confirm the response;

[0034] c. The reader R monitors the response status of each time slot. If no response is detected in a time slot, the corresponding tag is marked as "missing tag" again;

[0035] d. Reader R updates the tag status table based on the final recognition result, including adding the newly recognized unknown tag information and removing the confirmed missing tags to achieve a closed state loop.

[0036] Furthermore, the communication time calculation model used in step S3 includes three parts:

[0037] Write order numbers and fingerprints for n known tags, i.e. preloading overhead:

[0038] T1=n*C Write

[0039] Among them, C Write Indicates the communication time required to execute a C1G2 standard Write command;

[0040] Single tag activation is achieved through the Select+Query command sequence, that is, the tag polling overhead is known:

[0041] T2=n*(C Select +log2N+C Query )+n*C RN16

[0042] Among them, C Select 、C Query and C RN16 Respectively represent the communication time required to execute a C1G2 standard Select command, Query command and RN16 response;

[0043] The Select command matches both the order number and the fingerprint, i.e., the unknown tag filtering overhead:

[0044]

[0045] Add up these three parts of communication time to get the total communication time.

[0046] Furthermore, the tag status report generated in step S4 includes the following information:

[0047] a. Tag status statistics: Summarize the final data and classify and count the number and distribution of known tags, unknown tags, stopped tags, and missing tags;

[0048] b. Data visualization: intuitively display label distribution changes and recognition accuracy through trend charts;

[0049] c. Exception handling mechanism: Provides early warnings for frequently missing or abnormal labels and provides optimization suggestions;

[0050] d. System operation requirements: Based on the current tag status analysis, provide storage area capacity optimization suggestions and tag mapping table update strategies to maintain system recognition efficiency.

[0051] A continuous tag identification system based on COTS RFID of the present invention comprises:

[0052] Polling module: Efficiently polls known tags, quickly locates and activates target tags through pre-stored static identifiers, detects missing tags in real time and updates their status;

[0053] Screening module: uses fingerprint matching technology and combines random factors to generate dynamic identity tags to accurately distinguish known tags from unknown tags;

[0054] Verification module: performs secondary verification on the labels after preliminary screening and further eliminates potential missing labels through polling;

[0055] Optimization module: Based on information coding theory, it calculates the theoretical lower limit of the communication time of the tag identification process, analyzes the gap between the protocol execution efficiency and the theoretical optimal value, guides the adjustment of protocol parameters, and optimizes the polling strategy and filtering mechanism;

[0056] Report module: Integrate recognition data and generate visual reports, including label distribution, anomaly warning and system performance analysis.

[0057] 3. Beneficial effects

[0058] Compared with the existing known technologies, the technical solution provided by the present invention has the following significant effects:

[0059] (1) This invention proposes a continuous tag recognition method based on COTS RFID. This method addresses the challenges of tag dynamic changes in real-world deployment environments by implementing a three-stage efficient collaborative mechanism to achieve real-time dynamic updates of tag status and tag lists, significantly reducing communication overhead and improving overall system performance. To reduce communication time and improve recognition efficiency, the protocol fully utilizes the tag read sequence information, constructing an efficient filtering data structure and polling mechanism, avoiding redundant operations and collision time slots, and thus completing the recognition of unknown and missing tags in the shortest possible time.

[0060] (2) The present invention provides a continuous tag identification method based on COTS RFID. In the static identifier polling stage, the method uses a pre-written unique identifier to achieve rapid tag positioning, avoiding the additional overhead caused by dynamic updates. In the intelligent screening stage, the method combines a fingerprint verification mechanism to achieve accurate tag classification, significantly reducing the probability of misidentification. In the efficient verification stage, the method adopts a single time slot response strategy to ensure the effectiveness of each communication and avoid wasting empty time slots. Through theoretical analysis and experimental verification, it is proved that the communication time of the protocol is close to the theoretical lower limit, which is significantly better than existing protocols (such as OMTI, BUIT, CUD, FCS). In addition, the false alarm rate and missed alarm rate of the protocol are close to the theoretical values, ensuring the accuracy and reliability of the recognition results. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is a flow chart of the continuous identification method of mobile tags based on the COTS RFID system described in this embodiment.

[0062] Figure 2 Figure 2 shows an example of how the order number and fingerprint are stored in C1G2-compliant known and unknown tags at the beginning of a read cycle. DETAILED DESCRIPTION

[0063] Current research in RFID tag recognition technology focuses on three key areas: rapid tag authentication, tag status monitoring in dynamic environments, and optimizing system communication efficiency. In highly dynamic applications like logistics warehousing and smart retail, the constant flow of tags creates a dual challenge for real-time recognition technology: the need to instantly capture new tags entering the facility and accurately track those leaving.

[0064] Traditional solutions typically employ periodic full-disk scans or probabilistic sampling detection, but these methods suffer from inherent flaws—either wasting communication resources due to repeated identification of fixed tags or missing tags due to sampling bias. Existing technologies are generally limited to a single task: either focusing on new tag discovery while ignoring missing tag detection, or focusing on missing tag monitoring at the expense of new tag recognition efficiency. This functional fragmentation makes it difficult to balance real-time performance and resource consumption in practical deployments. Furthermore, commercial-off-the-shelf (COTS) RFID devices are constrained by the C1G2 standard and cannot support protocols requiring complex tag-side computation, making many theoretical optimization methods difficult to implement.

[0065] To address these technical bottlenecks, the present invention proposes a continuous tag identification method based on a COTS RFID system. By preloading static order numbers and fingerprint data, this method achieves near-theoretical optimal performance on commercial off-the-shelf (COTS) equipment. Specifically, this method leverages the selective reading mechanism of the C1G2 standard, simulating a dynamic counter function through static order numbers to achieve efficient polling and missing detection of known tags. Simultaneously, an optimized filtering structure is employed to isolate unknown tags in a manner that minimizes communication overhead. By merging Select and Query command operations, this protocol effectively reduces the additional overhead introduced by redundant fields, thereby significantly improving system efficiency while ensuring recognition accuracy.

[0066] In summary, the present invention is a dynamic tag identification method for commercial equipment. Its innovations are reflected in: 1) using pre-configured static identifiers instead of dynamic counters, and realizing selective reading through standard Select commands; 2) designing a two-stage filtering mechanism, first separating known tag groups through order number matching, and then using fingerprint verification to exclude pseudo-unknown tags; 3) optimizing command sequence combinations to eliminate redundant fields in standard protocols.

[0067] Experimental data demonstrates that this method significantly reduces average recognition latency while maintaining C1G2 standard compatibility and supporting a high dynamic update rate. Compared to existing C1G2-compatible protocols (such as P2P+BF and GTI+BF), this method significantly reduces communication time while maintaining high recognition accuracy. This method demonstrates excellent adaptability and real-time performance in dynamically changing RFID environments, providing an efficient and reliable solution for practical applications such as logistics tracking, smart warehousing, and retail management.

[0068] The continuous tag identification method based on the COTS RFID system disclosed in the present invention is further described in detail below with reference to specific embodiments.

[0069] Example 1

[0070] A continuous tag recognition method based on COTS RFID in this embodiment, denoted as CRSP algorithm, includes the following steps:

[0071] Step S1, such as Figure 2 As shown, at the beginning of the reading cycle, the order number and fingerprint are stored in the C1G2 compatible version of the known tag and the unknown tag, wherein the known tag is a tag that has been registered in the system database and assigned a unique static identifier, and the unknown tag is a new tag that has not been registered in the system or assigned a static identifier.

[0072] In step S2, an identity benchmark is established by pre-storing a unique static identifier for each known tag in the user storage area. Standard commands (Select / Query) in the C1G2 protocol are used to selectively poll known tags to detect missing tags within the current identification range. Missing tags are registered tags that have not responded within the current cycle, which may be caused by leaving the read / write range or signal problems. A fingerprint-based optimized filtering structure is then constructed and combined with the fingerprint filtering structure to accurately separate and identify unknown tags. Based on the polling and filtering results, the remaining tags are further efficiently confirmed. This efficient confirmation mechanism optimizes the identification process, ultimately ensuring the complete identification of all present tags, improving tag recognition accuracy and system stability.

[0073] In this step, the communication overhead is significantly reduced through the following three-phase operation:

[0074] In the first phase (static identifier polling phase for known tags), the reader screens known tags through a static identifier-driven mask polling mechanism, effectively identifying missing tags and avoiding redundant communications caused by unresponsive tags.

[0075] In the second stage (intelligent screening stage), the reader builds a fingerprint filtering structure based on the static identifier index, and guides the tag to generate a local fingerprint by broadcasting a random factor, thereby achieving low-cost screening of unknown tags.

[0076] In the third phase (efficient verification phase), the reader again performs verification polling on the remaining tags in a one-to-one manner, avoiding the waste of collision time slots generated in the traditional frame structure and further reducing communication overhead.

[0077] Through the coordinated execution of the above three stages, this method can achieve efficient tag identification and missing detection in a dynamic RFID environment, while maintaining the communication cost at a low level and improving the overall performance of the system.

[0078] Specifically, for the static identifier polling phase of a known tag, the following steps are performed:

[0079] a. The reader R maintains a tag list L, assigns a unique static identifier to each known tag, and writes it to its user storage area;

[0080] b. Each known tag t sends a predefined response signal "1" in the corresponding time slot according to the stored static identifier;

[0081] c. The reader R monitors the time slot response in real time. If a "1" signal is not received in a time slot, the tag corresponding to the time slot is marked as "missing";

[0082] d. The reader R constructs an n-bit response array S based on the detection results, identifies the response status of each tag, and broadcasts it to all tags;

[0083] e. Each tag updates its local state variable (such as a register or counter) according to the response array S and records whether its own response state is normal.

[0084] For the intelligent screening stage, an optimized filtering structure is constructed to efficiently separate and label unknown tags with low communication overhead. The following steps are performed:

[0085] a. The reader R generates and broadcasts a random number v and fingerprint length parameter D as input for the tag to calculate the identity fingerprint;

[0086] b. After receiving the random number v, each tag t calculates a fingerprint value h(t) based on its own ID and the random number v, for example, by generating a fixed-length bit string using a hash function;

[0087] c. The reader R constructs a filtering data structure F containing m units for the fingerprints h(t) of all known tags, where F[i] stores the fingerprint of the tag corresponding to the i-th static identifier;

[0088] d. Each tag t searches for its corresponding position in F according to its local static identifier. If the comparison result is consistent with the local fingerprint, it is marked as a "known tag", otherwise it is marked as an "unknown tag".

[0089] For the efficient verification phase, further polling of the remaining tags to identify more missing tags and ensure the accuracy of the identification of the currently present tags, the following steps are performed:

[0090] a. The reader R restarts a frame consisting of k time slots, where k is the number of known tags to be verified, and each time slot is synchronized with the static identifier of the tag;

[0091] b. Each tag t sends a "1" signal in the corresponding time slot according to its static identifier to confirm the response;

[0092] c. The reader R monitors the response status of each time slot. If no response is detected in a time slot, the corresponding tag is marked as "missing tag" again;

[0093] d. Reader R updates the tag status table based on the final recognition result, including adding the newly recognized unknown tag information and removing the confirmed missing tags to achieve a closed state loop.

[0094] Step S3: By accurately calculating the theoretical lower limit of the CRSP protocol's communication time and guiding the optimization design of the protocol parameters based on this theoretical lower limit, the tag reading order and filtering mechanism are systematically improved, ultimately improving the overall recognition efficiency of the protocol.

[0095] In commercial off-the-shelf (COTS) RFID systems, there is a theoretical lower bound on the communication time for the continuous tag identification problem. This lower bound is based on information encoding and probability analysis. The specific derivation process is as follows:

[0096] Define the lower limit of communication time: Let T LB is the theoretical lower limit of communication time, T ID The unit of time required to transmit a 96-bit tag ID. According to information coding theory, the lower limit of communication time is given by the following formula:

[0097]

[0098] Among them, Γ is a parameter related to the number of labels, false positive rate and false negative rate.

[0099] The specific expression of parameter Γ is:

[0100]

[0101] Where n is the total number of known labels, represents the number of tags that actually exist, including resident tags (known tags that are still within the reader range during the current cycle, indicating that they can continue to be identified) and missing tags that share the reading order with unknown tags. ε and ξ represent the false alarm rate and missed alarm rate, respectively. Under this condition, the minimum communication time required to separate unknown tags is:

[0102]

[0103] For the task of detecting missing labels, the minimum communication time required is:

[0104] n[1+ξlog2ξ+(1-ξ)log2(1-ξ)]

[0105] Adding the two parts of the communication time gives Γ, the final expression of the theoretical lower limit: Substituting Γ into the formula for the lower limit of the communication time, we get:

[0106]

[0107] In COTS RFID systems, the lower bound on the communication time for continuous tag identification is determined by the system implementation and protocol parameters. Based on the actual overhead of C1G2 standard commands, this embodiment proposes the following communication time analysis model:

[0108] The total communication time consists of three parts:

[0109] Write order numbers and fingerprints for n known tags, i.e. preloading overhead:

[0110] T1=n*C Write

[0111] Among them C Write Indicates the communication time required to execute a C1G2 standard Write command, using a Select+Query command sequence to activate a single tag, i.e., the known tag polling overhead:

[0112] T2=n*(C Select +log2N+C Query )+n*C RN16

[0113] Among them C Select 、C Query and C RN16 The communication times required to execute a C1G2 standard Select command, Query command, and RN16 response, respectively. The Select command matches both the order number and the fingerprint, i.e., the unknown tag filtering overhead:

[0114]

[0115] Adding up these three parts of communication time gives the total communication time:

[0116]

[0117] The theoretical lower bound is derived based on information coding and probabilistic analysis, ensuring that the communication time required by any protocol for solving the problem of continuous tag identification is at least this lower bound. Using this theoretical lower bound as a guide, the protocol optimizes the tag reading sequence and filtering mechanism to bring its communication time closer to the theoretical optimal value, significantly improving tag recognition efficiency while ensuring reliable identification.

[0118] In step S4, the reader generates a detailed tag status report based on the data results, providing users with intuitive real-time monitoring data and supporting dynamic adjustments to the RFID system. The report covers the following key information:

[0119] a. Tag status statistics: Summarize the final data and classify and count the number and distribution of known tags, unknown tags, stopped tags, and missing tags.

[0120] b. Data visualization: Trend charts are used to intuitively display changes in label distribution and recognition accuracy, making it easier for users to understand the system's operating status.

[0121] c. Exception handling mechanism: Provides early warnings for frequently missing or abnormal tags, and provides optimization suggestions such as polling frequency adjustment to improve system performance.

[0122] d. System operation requirements: Based on the current tag status analysis, provide storage area capacity optimization suggestions and tag mapping table update strategies to maintain system recognition efficiency.

[0123] The performance of the method of this embodiment is verified by the following methods:

[0124] A realistic RFID experimental environment was constructed using commercial off-the-shelf (COTS) equipment. The test platform employed the Impinj R420 reader and ALN-9662 tags to simulate three typical scenarios of tag dynamics in real-world applications: tag population growth, decrease, and steady state. Comparative testing with traditional protocols (OMTI, BUIT, CUD, and FCS) and mainstream C1G2-compatible solutions (P2P+BF and GTI+BF) conducted a comprehensive evaluation based on communication efficiency, recognition accuracy, and system stability.

[0125] Experimental data shows that this solution has significant advantages in real hardware environments:

[0126] Communication efficiency: Average communication time is reduced by 50.3%-62.8% compared to the comparison protocol, and stable performance is maintained even in scenarios where tags are dynamically changing.

[0127] Identification accuracy: The false alarm rate is controlled below 0.8%, and the missed alarm rate does not exceed 1.2%, which is significantly better than P2P+BF (false alarm 3.5%, missed alarm 4.1%).

[0128] System stability: During an 8-hour continuous stress test, the recognition success rate reached 99.4%, with a fluctuation range of less than 0.3%.

[0129] Especially in scenarios where tags are rapidly entering and exiting (20-30 changes per minute), this embodiment can still maintain a real-time recognition rate of over 95%, validating its deployment value in real-world commercial environments. Experimental data fully demonstrates that this solution not only meets theoretical expectations but also achieves breakthroughs in both performance and reliability on actual C1G2 devices.

[0130] This method is applicable to a variety of dynamic change scenarios, including but not limited to:

[0131] Cargo distribution chain scenarios where the number of tags continues to increase;

[0132] Warehouse outbound operation scenarios where the number of labels decreases periodically;

[0133] Production pipeline scenarios where the label set remains relatively stable;

[0134] In the above scenarios, this method shows lower communication time and higher recognition accuracy than traditional protocols, especially showing good scalability in high-density tag environments.

[0135] The false alarm rate and missed alarm rate of this method are close to the theoretical lower limit, and the estimated expression is:

[0136] False alarm rate ε: It is determined by the length of the tag fingerprint and the probability of hash function collision. It is a threshold that can be set by the system and is generally set to ε≈10 -3 ~10 -6 ;

[0137] Missing rate ξ: satisfies ξ = ε·r·(1-α), where r represents the conflict probability factor in the filtering structure, and α is the tag presence rate (i.e., the ratio of the number of tags present to the total number of tags).

[0138] Example 2

[0139] The continuous tag identification method based on the COTS RFID system disclosed in the present invention can be run in a processor or stored in a computer-readable storage medium. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, which can store information through any method or technology. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (Flash Memory) or other memory technology, read-only compact disc memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic tape, disk storage device or other magnetic storage device, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. According to the definition in this embodiment, computer-readable media does not include temporary computer-readable media, such as modulated data signals and carrier waves.

[0140] These computer programs can also be loaded into a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to generate a computer-implemented process, thereby providing instructions executed on the computer or other programmable device to provide steps for implementing the functions specified in the flowchart or block diagram. The functions corresponding to different steps can be implemented by different modules.

[0141] In this embodiment, such a system is provided. The system can be called a continuous tag identification system based on a COTS RFID system. The system includes:

[0142] The polling module efficiently polls known tags, quickly locating and activating target tags using pre-stored static identifiers. If a tag is detected as missing in real time, it is marked as missing. Its core function is to reduce redundant communication, optimize polling sequences, and ensure the system quickly responds to changes in tags in a dynamic environment, providing an accurate data foundation for subsequent processing.

[0143] The screening module uses fingerprint matching technology combined with random factors to generate dynamic identity tags, accurately distinguishing between known and unknown tags. Its function is to reduce the false recognition rate, avoid invalid communication, and improve the recognition efficiency of newly added tags, ensuring the system can maintain stable operation in high-density tag environments.

[0144] Verification Module: This module performs secondary verification on tags after initial screening, further eliminating potential missing tags through precise polling and compensating for misjudgments caused by signal interference or time slot conflicts. Its core function is to improve recognition reliability, ensure the accuracy of the final tag status, and provide more stable data support for the system.

[0145] Optimization Module: Based on information coding theory, this module calculates the theoretical lower bound of communication time during tag identification and analyzes the gap between protocol execution efficiency and the theoretical optimal value. Its core function is to use mathematical models to guide protocol parameter adjustments, optimize polling strategies and filtering mechanisms, and continuously bring system performance closer to the theoretical optimal value, significantly improving overall identification efficiency.

[0146] Reporting module: This module integrates identification data and generates visual reports, including tag distribution, anomaly warnings, and system performance analysis. Its purpose is to help users intuitively understand tag status changes, optimize system operation strategies, and provide data support for decision-making, further enhancing the intelligent management level of the RFID system.

Claims

1. A continuous tag identification method based on COTS RFID, characterized in that: The following steps are involved: Step S1: At the beginning of the reading cycle, the order number and fingerprint are stored in the C1G2 compatible version of the known tag and the unknown tag; Step S2: Pre-store a unique static identifier for each known tag, selectively poll the known tags, detect missing tags within the current identification range, and build a fingerprint-based filtering structure to separate and identify unknown tags; based on the polling and filtering results, further confirm the remaining tags; Step S3: Calculate the theoretical lower limit of the communication time using the communication time calculation model, and guide the optimization of protocol parameters based on the theoretical lower limit value to improve the tag reading order and filtering mechanism; Step S4: The reader generates a tag status report based on the data results, provides real-time monitoring data to the user, and supports dynamic adjustment of the RFID system.

2. The COTS RFID-based continuous tag identification method according to claim 1, characterized in that: The known tags are tags that have been registered in the system database and assigned a unique static identifier; the unknown tags are new tags that have not been registered in the system or assigned a static identifier; the missing tags are tags that have been registered in the system database but have not responded within the identification period.

3. The COTS RFID-based continuous tag identification method according to claim 1 or 2, characterized in that: Step S2 includes three operation stages: In the first phase, the reader screens known tags through a static identifier-driven masked polling mechanism to identify missing tags and avoid redundant communications caused by unresponsive tags; In the second stage, the reader builds a fingerprint filtering structure based on the static identifier index, and guides the tag to generate a local fingerprint by broadcasting a random factor, thereby filtering out unknown tags. In the third phase, the reader again performs verification polling on the remaining tags in a one-to-one manner to avoid the waste of collision time slots generated in the traditional frame structure.

4. The COTS RFID-based continuous tag identification method according to claim 3, characterized in that: In the first phase, perform the following steps: a. The reader R maintains a tag list L, assigns a unique static identifier to each known tag, and writes it to the user storage area; b. Each known tag t sends a predefined response signal "1" in the corresponding time slot according to the stored static identifier; c. The reader R monitors the time slot response in real time. If a time slot does not receive a "1" signal, the tag corresponding to the time slot is marked as "missing"; d. The reader R constructs an n-bit response array S based on the detection results, identifies the response status of each tag, and broadcasts it to all tags; e. Each tag updates its local state variable according to the response array S and records whether its own response status is normal.

5. The COTS RFID-based continuous tag identification method according to claim 4, characterized in that: In the second phase, perform the following steps: a. The reader R generates and broadcasts a random number v and fingerprint length parameter D as input for the tag to calculate the identity fingerprint; b. After each tag t receives the random number v, it calculates the fingerprint value h(t) based on its own ID and the random number v; c. The reader R constructs a filtering data structure F containing m units for the fingerprints h(t) of all known tags, where F[i] stores the fingerprint of the tag corresponding to the i-th static identifier; d. Each tag t is searched for its corresponding position in F according to its local static identifier. If the comparison result is consistent with the local fingerprint, it is marked as a "known tag", otherwise it is marked as an "unknown tag".

6. The COTS RFID-based continuous tag identification method according to claim 5, characterized in that: In the third phase, perform the following steps: a. The reader R restarts a frame consisting of k time slots, where k is the number of known tags to be verified, and each time slot is synchronized with the static identifier of the tag; b. Each tag t sends a "1" signal in the corresponding time slot to confirm the response according to its static identifier; c. The reader R monitors the response status of each time slot. If no response is detected within a time slot, the corresponding tag is marked as "missing tag" again; d. Reader R updates the tag status table based on the final recognition result, including adding the newly recognized unknown tag information and removing the confirmed missing tags to achieve a closed state loop.

7. The COTS RFID-based continuous tag identification method according to claim 6, characterized in that: The communication time calculation model used in step S3 includes three parts: Write order numbers and fingerprints for n known tags, i.e. preloading overhead: T1=n*C Write Among them, C Write Indicates the communication time required to execute a C1G2 standard Write command; Single tag activation is achieved through the Select+Query command sequence, that is, the tag polling overhead is known: T2=n*(C Select +log2N+C Quert )+n*C RN16 Among them, C Select 、C Query and C RN16 Respectively represent the communication time required to execute a C1G2 standard Select command, Query command and RN16 response; The Select command matches both the order number and the fingerprint, i.e., the unknown tag filtering overhead: Add up these three parts of communication time to get the total communication time.

8. The COTS RFID-based continuous tag identification method according to claim 7, characterized in that: The tag status report generated in step S4 includes the following information: a. Tag status statistics: Summarize the final data and classify and count the number and distribution of known tags, unknown tags, stopped tags, and missing tags; b. Data visualization: intuitively display label distribution changes and recognition accuracy through trend charts; c. Exception handling mechanism: Provides early warnings for frequently missing or abnormal labels and provides optimization suggestions; d. System operation requirements: Based on the current tag status analysis, provide storage area capacity optimization suggestions and tag mapping table update strategies to maintain system recognition efficiency.

9. A continuous tag identification system based on COTS RFID, characterized in that: include: Polling module: Efficiently polls known tags, quickly locates and activates target tags through pre-stored static identifiers, detects missing tags in real time and updates their status; Screening module: uses fingerprint matching technology and combines random factors to generate dynamic identity tags to accurately distinguish known tags from unknown tags; Verification module: performs secondary verification on the labels after preliminary screening and further eliminates potential missing labels through polling; Optimization module: Based on information coding theory, it calculates the theoretical lower limit of the communication time of the tag identification process, analyzes the gap between the protocol execution efficiency and the theoretical optimal value, guides the adjustment of protocol parameters, and optimizes the polling strategy and filtering mechanism; Report module: Integrate recognition data and generate visual reports, including label distribution, anomaly warning and system performance analysis.

Images