Intelligent file cabinet management method based on RFID technology

Abstract

The invention relates to the technical field of archive management, and discloses an RFID technology-based intelligent archive cabinet management method, which comprises the following steps of S1, inputting label information and archive information in a system, performing encryption storage through a block chain technology, arranging a sensor in an archive cabinet, acquiring environment data, and associating the environment data with RFID label information in real time; s2, an RFID reader-writer, a sensor and a camera are installed, the working states of the RFID reader-writer, the sensor and the camera are debugged to collect data in various environments, a self-adaptive signal enhancement mechanism is introduced, and the power and the frequency of the RFID reader-writer are adjusted according to the real-time environment. By combining the block chain technology and the digital signature technology, the system ensures that file data cannot be tampered and the operation safety is ensured, the introduction of the block chain provides a highly transparent and safe operation platform for file management, and the overall safety of the system is significantly improved. An intelligent fault prediction and label failure detection mechanism can identify potential problems in advance and prevent faults.

Description

Technical Field

[0001] The present invention relates to the technical field of file management, and in particular to an intelligent file cabinet management method based on RFID technology. Background Art

[0002] Archives management, also known as archives work, is the general term for all business operations of the archives department that directly manages archive entities and archives information and provides utilization services.

[0003] With the continuous development of information technology, archive management has gradually transitioned from traditional manual operations to intelligent management systems. Traditional archive management methods usually rely on manual registration, manual access and regular manual inventory. These methods are time-consuming and labor-intensive, and are prone to human errors, resulting in inefficient archive management and insufficient accuracy. In addition, traditional archive management systems have many problems in dealing with data security and privacy protection, especially when facing the threat of data tampering and illegal access, and lack effective protection mechanisms.

[0004] In recent years, RFID technology has been widely used in archive management. With its advantages of automatic identification and long-distance reading, RFID technology has significantly improved the efficiency and accuracy of archive management. However, the existing archive management system based on RFID technology still faces important challenges. For example, RFID technology can automatically read and record the status of archives, but due to the lack of strong security protection during data transmission and storage, archive data is still subject to tampering or leakage. In addition, existing systems are mostly statically configured and cannot adapt to dynamically changing environments and business needs. When the system expands or the usage environment changes, manual intervention is required and lacks adaptive capabilities.

[0005] On the other hand, the existing archive management system usually relies on post-processing when dealing with abnormal situations, such as discovering the loss or damage of archives through regular inventory. Its passive management method is slow to respond and can easily lead to serious loss or damage of archives that cannot be recovered in time. In addition, the traditional system lacks real-time monitoring and intelligent analysis capabilities of environmental changes, making it difficult to foresee and prevent abnormal situations.

[0006] Therefore, those skilled in the art provide an intelligent filing cabinet management method based on RFID technology to solve the problems raised in the above background technology. Summary of the invention

[0007] In view of the deficiencies of the prior art, the present invention provides an intelligent filing cabinet management method based on RFID technology to solve the problems raised in the above background technology.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: A smart filing cabinet management method based on RFID technology, comprising the following steps: S1, entering tag information and file information into the system, and encrypting and storing them through blockchain technology, and at the same time, arranging sensors in the filing cabinet to collect environmental data and associate them with RFID tag information in real time;

[0009] S2. Install RFID readers, sensors and cameras, and debug their working status to collect data in various environments. In addition, an adaptive signal enhancement mechanism is introduced to adjust the power and frequency of the RFID readers according to the real-time environment.

[0010] S3. Put the file into the smart filing cabinet. The system automatically reads the RFID tag and records the storage location of the file. It also collects and records environmental data. The information is encrypted and stored in the blockchain.

[0011] S4. Query through the archive management system. The system can accurately locate and verify the archive status based on RFID and sensor data combined with camera images. When taking out or returning archives, the system automatically updates the archive status and records the operation information in combination with blockchain.

[0012] S5. Regularly trigger RFID readers and sensors to scan the entire disk. At the same time, use camera images to verify the status of the file. The system detects anomalies through big data analysis and machine learning algorithms, generates inventory reports, and marks abnormal situations.

[0013] S6. By analyzing the file usage patterns and environmental data, the system automatically identifies abnormal operations and issues real-time alarms to prevent data loss or file damage. At the same time, the user operation permissions are strictly managed through blockchain technology and recorded in the blockchain ledger, and blockchain data is audited regularly;

[0014] S7. The system automatically adjusts the signal parameters of the RFID reader according to environmental changes, and analyzes the operating status of the equipment through machine learning algorithms, predicts equipment failure or tag failure, and performs maintenance or replacement in advance;

[0015] S8. Utilize the collected multi-dimensional data to optimize the archive management strategy through big data analysis, adjust the storage layout, and regularly upgrade the system software and expand the hardware according to the development of new technologies and changes in management needs.

[0016] Preferably, in the step S1, in the real-time association and encrypted storage of the environmental data and the RFID tag information, the real-time association formula of the environmental data and the RFID tag information is as follows:

[0017] E(t) represents the environmental data vector collected at time t;

[0018] R(t) represents the RFID tag information vector collected at time t;

[0019] α i is the weight coefficient of different environmental parameters;

[0020] β j is the weight coefficient of different RFID data parameters;

[0021] S(t) represents the comprehensive index of archive status calculated at time t;

[0022]

[0023] Among them, m is the dimension number of environmental data, and n is the dimension number of RFID data;

[0024] To ensure the integrity and non-repudiation of data, digital signature technology is used, and its algorithm formula is:

[0025] SK is the private key, used for signing; PK is the public key, used for verifying the signature; Sig is the signature result;

[0026] The digital signature generation process is expressed as:

[0027] Sig=Sign(SK,H(D)),

[0028] The transaction data in the blockchain will include Sig and H(D) and will be verified by the public key PK:

[0029] Verify(PK,H(D),Sig)=True or False,

[0030] Successful verification indicates that the integrity and legitimacy of the data are guaranteed.

[0031] Preferably, in the step S2, the adaptive signal enhancement mechanism and the sensor data fusion and tuning formula are used to optimize the power and frequency adjustment of the RFID reader to adapt to different environments, and dynamically adjust in combination with the sensor data;

[0032] The adaptive signal enhancement mechanism formula:

[0033] P(t) is the output power of the RFID reader at time t; f(t) is the operating frequency of the RFID reader at time t; E(t) is the environmental data vector collected at time t; R(t) is the RFID signal strength at time t; γ(t) is the environmental impact factor at time t;

[0034] Calculation of environmental impact factors: Calculate the environmental impact factor γ(t) based on the environmental data E(t):

[0035]

[0036] Among them, α i is the weight of each environmental parameter, and m is the number of dimensions of environmental data;

[0037] Power adjustment formula: Adjust the output power P(t) of the RFID reader according to the size of γ(t):

[0038] P(t+1)=P(t)·(1+k1·γ(t)),

[0039] Wherein, k1 is the adjustment coefficient, which is used to control the adjustment amplitude of the power, and P(t+1) is the output power at time t+1;

[0040] Frequency adjustment formula: Adjust the operating frequency f(t) according to the signal strength R(t) and the environmental impact factor γ(t):

[0041]

[0042] Wherein, k2 is the frequency adjustment coefficient.

[0043] Preferably, the sensor data fusion and dynamic tuning formula is to ensure the best performance of the RFID reader in a complex environment. The system fuses the sensor data and performs dynamic tuning. The algorithm formula is:

[0044] S(t) is the weighted fusion result of sensor data; θ(t) is the RFID performance tuning parameter set at time t; L(θ(t)) is the performance loss function;

[0045] Sensor data fusion formula: Combining the data from different sensors, we get the fused environmental state index S(t):

[0046]

[0047] Among them, w j is the weight of each sensor data, D j (t) is the data of sensor j at time t, n is the number of sensors;

[0048] Dynamic tuning formula: Use the gradient descent method to optimize the tuning parameters θ(t) of the RFID system:

[0049]

[0050] Where η is the learning rate, is the gradient of the loss function L with respect to the parameter θ(t).

[0051] Preferably, in the step S4, the blockchain recording formulas for archive query, positioning, status verification and data are used to optimize the accuracy of archive query, the accuracy of positioning, the verification of status and the security of blockchain recording;

[0052] The file query and positioning algorithm formula: File query and positioning require comprehensive analysis based on RFID tag data, sensor data and camera image information:

[0053] Q(t) is the archive query request at time t, R(t) is the response data read by the RFID tag, C(t) is the image data obtained by the camera, and L(t) is the archive location information finally located by the system;

[0054] File location formula: File location is performed by integrating RFID data and image data:

[0055] L(t)=w1·R(t)+w2·Image_Process(C(t)),

[0056] Among them, w1 and w2 are the weight coefficients of RFID data and image data,

[0057] Image_Process(C(t)) represents the extraction of position information after image data processing.

[0058] Status verification algorithm formula: After locating the file, the system needs to verify the status of the file, whether it is complete and in the correct location:

[0059] S expected is the expected status of the archive recorded in the system, S actual (t) is the actual status of the archive acquired through sensor and image data at time t;

[0060] Status Verification Formula: Profile status verification is performed by comparing the expected status with the actual status:

[0061] ΔS(t)=S expected -S actual (t),

[0062] If ΔS(t) is less than the preset tolerance ∈, the state verification is considered to have passed:

[0063]

[0064] Preferably, in the step S5, full disk scanning, camera image verification, anomaly detection and big data analysis can achieve comprehensive monitoring of the archive status and detection of abnormal situations through the fusion and intelligent analysis of multi-source data;

[0065] Anomaly detection identifies abnormal status of files by performing big data analysis on full disk scan and image verification data:

[0066] D(t) is the set of all archive status data scanned at time t, including S i (t) and image verification results, μ D is the expected value of the data set D(t), σ D is the standard deviation of the data set D(t);

[0067] Anomaly detection formula: Use the z-score method to detect anomalies in the status data of each file:

[0068] Z i (t) is the z-score of the i-th file status, and the z-score calculation formula is:

[0069]

[0070] If | Z i (t)|>Z threshold , then the file status is considered abnormal:

[0071]

[0072] Preferably, in the step S6, the identification of abnormal operations, the triggering of real-time alarms, the management of user operation rights, and the audit and verification of blockchain data are performed, wherein the identification of abnormal operations detects abnormal behaviors by analyzing changes in user operation patterns and environmental data:

[0073] U(t) is the user's operation data at time t, H(t) is the environmental data change vector at time t, O(t) is the comprehensive operation behavior vector, which represents the combined information of operation and environment at time t, and θ is the standard parameter set of the operation mode;

[0074] Operation behavior vector formula: Combine user operation data U(t) with environmental data change H(t) to form an operation behavior vector:

[0075] O(t)=U(t)+λ·H(t),

[0076] Among them, λ is the weight coefficient;

[0077] Anomaly identification formula: Assumption O expected,i The expected value of the behavior vector for normal operation is used to determine whether the operation is abnormal using the distance measurement method:

[0078]

[0079] If Dist(O(t),O expected) exceeds the set threshold ∈, it is judged as an abnormal operation:

[0080]

[0081] Real-time alarm triggering algorithm formula: If the system detects abnormal operation, the alarm should be triggered in time:

[0082] A(t) is the anomaly detection result at time t, and τ(t) is the time when the alarm is triggered at time t;

[0083] Real-time alarm formula: If the anomaly detection result is true, the system triggers an alarm:

[0084]

[0085] Preferably, in the step S7, the RFID reader signal is used for adaptive adjustment, prediction of equipment failure and early detection of tag failure. The RFID reader signal adaptive adjustment algorithm formula automatically optimizes its working parameters according to changes in environmental conditions:

[0086] P(t) is the output power of the RFID reader at time t, f(t) is the operating frequency of the RFID reader at time t, and E(t) is the environmental impact factor at time t;

[0087] Output power adjustment formula: Adjust the output power of the RFID reader according to the environmental impact factor E(t):

[0088] P(t+1)=P(t)·(1+k p E(t)),

[0089] Among them, k p is the power adjustment coefficient, which controls the sensitivity of power adjustment;

[0090] Working frequency adjustment formula: Adjust the working frequency of the RFID reader according to environmental changes E(t) and signal quality Q(t):

[0091]

[0092] Among them, k f is the frequency adjustment coefficient, ∈ is a small constant to avoid the denominator being zero, and Q(t) represents the signal quality indicator.

[0093] Preferably, in the step S8, the multidimensional data collection and aggregation algorithm formula is used for the collection and analysis of multidimensional data and the optimization of archive management strategies and the automatic upgrade and expansion of the system:

[0094] D i (t) is the data vector collected by the i-th sensor or device at time t, Wi is the weight of the i-th data source, M(t) is the multidimensional data aggregation result at time t;

[0095] Data aggregation formula: aggregate data from different sources according to weights;

[0096]

[0097] Where n is the number of data sources.

[0098] Preferably, in the step S8, the system self-learning and model self-adaptation formula is to cope with the ever-changing needs, and the system must have self-learning and self-adaptation capabilities:

[0099] θ(t) is the set of model parameters at time t, and L(θ(t)) is the loss function of the model, which measures the performance of the model;

[0100] Self-learning update formula: Use adaptive learning rate to perform self-learning update of model parameters:

[0101]

[0102] Among them, the learning rate η(t) is dynamically adjusted according to the model performance to accelerate the learning process:

[0103] η(t+1)=η(t)·(1+κ·ΔL(t)),

[0104] Among them, κ is the learning rate adjustment coefficient, and ΔL(t) is the rate of change of the loss function.

[0105] The present invention provides an intelligent filing cabinet management method based on RFID technology. It has the following beneficial effects:

[0106] 1. The present invention combines blockchain technology and digital signature technology to ensure that the archive data cannot be tampered with and that operations are safe. The introduction of blockchain provides a highly transparent and secure operating platform for archive management, significantly improving the overall security of the system. In addition, the intelligent fault prediction and tag failure detection mechanism can identify potential problems in advance and prevent faults from occurring. Its forward-looking maintenance method is innovative.

[0107] 2. The present invention introduces a self-learning algorithm and a dynamic resource allocation mechanism, which enables the system to automatically adjust according to environmental changes and business needs without relying on human intervention. Its intelligent adaptability and scalability enable the system to operate flexibly in different usage scenarios and can automatically upgrade to adapt to future technological advances.

[0108] 3. The present invention realizes all-round intelligent monitoring of the archive status through full disk scanning, anomaly detection and real-time alarm functions. Big data analysis and machine learning algorithms are used to identify complex abnormal patterns and provide detailed abnormality reports. Its high level of intelligent abnormality management improves the system's early warning capability and response speed, and innovatively solves the problem of passive handling of abnormal situations in traditional archive management systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0109] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0110] In order to make the technical personnel in the technical field understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in combination with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a partial embodiment of the present invention, not a complete embodiment. Based on the embodiment of the present invention, other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present invention.

[0111] The present invention is described in detail below in conjunction with the accompanying drawings:

[0112] Example:

[0113] Please see attached Figure 1 ,The embodiment of the present invention provides a smart filing cabinet management method based on RFID technology, including the following steps: S1, entering the tag information and the file information into the system, and encrypted and stored through the blockchain technology, and at the same time, arranging sensors in the filing cabinet to collect environmental data, and associating it with the RFID tag information in real time;

[0114] S2. Install RFID readers, sensors and cameras, and debug their working status to collect data in various environments. In addition, an adaptive signal enhancement mechanism is introduced to adjust the power and frequency of the RFID readers according to the real-time environment.

[0115] S3. Put the file into the smart filing cabinet. The system automatically reads the RFID tag and records the storage location of the file. It also collects and records environmental data. The information is encrypted and stored in the blockchain.

[0116] S4. Query through the archive management system. The system can accurately locate and verify the archive status based on RFID and sensor data combined with camera images. When taking out or returning archives, the system automatically updates the archive status and records the operation information in combination with blockchain.

[0117] S5. Regularly trigger RFID readers and sensors to scan the entire disk. At the same time, use camera images to verify the status of the file. The system detects anomalies through big data analysis and machine learning algorithms, generates inventory reports, and marks abnormal situations.

[0118] S6. By analyzing the file usage patterns and environmental data, the system automatically identifies abnormal operations and issues real-time alarms to prevent data loss or file damage. At the same time, the user operation permissions are strictly managed through blockchain technology and recorded in the blockchain ledger, and blockchain data is audited regularly;

[0119] S7. The system automatically adjusts the signal parameters of the RFID reader according to environmental changes, and analyzes the operating status of the equipment through machine learning algorithms, predicts equipment failure or tag failure, and performs maintenance or replacement in advance;

[0120] S8. Utilize the collected multi-dimensional data to optimize the archive management strategy through big data analysis, adjust the storage layout, and regularly upgrade the system software and expand the hardware according to the development of new technologies and changes in management needs.

[0121] The benefit of S1 is to ensure that data cannot be tampered with, prevent unauthorized access and tampering, help monitor the storage environment of archives, and prevent environmental factors from damaging archives; the benefit of S2 is to ensure that label information can be read stably and accurately in various environments, ensure comprehensive and real-time archive status monitoring, and enhance the breadth and depth of data collection; the benefit of S3 is to reduce manual intervention, improve the efficiency of archive management, and help maintain the integrity and security of archives; the benefit of S4 is to ensure the accuracy of archive management, reduce manual operation errors, and improve the transparency and security of management; the benefit of S5 is to ensure comprehensive monitoring of archives, detect problems in a timely manner, and improve the predictability and response speed of management; the benefit of S6 is to prevent data loss or archive damage, and ensure the transparency and traceability of operations; the benefit of S7 is to carry out maintenance or replacement in advance to reduce the risk of unexpected downtime and data loss; the benefit of S8 is to improve management efficiency and maintain the advancement and adaptability of the system.

[0122] In step S1, in the real-time association and encrypted storage of environmental data and RFID tag information, the real-time association formula of environmental data and RFID tag information is as follows:

[0123] E(t) represents the environmental data vector collected at time t;

[0124] R(t) represents the RFID tag information vector collected at time t;

[0125] α i is the weight coefficient of different environmental parameters;

[0126] β j is the weight coefficient of different RFID data parameters;

[0127] S(t) represents the comprehensive index of archive status calculated at time t;

[0128]

[0129] Among them, m is the dimension number of environmental data, and n is the dimension number of RFID data;

[0130] To ensure the integrity and non-repudiation of data, digital signature technology is used, and its algorithm formula is:

[0131] SK is the private key, used for signing; PK is the public key, used for verifying the signature; Sig is the signature result;

[0132] The digital signature generation process is expressed as:

[0133] Sig=Sign(SK,H(D)),

[0134] The transaction data in the blockchain will include Sig and H(D) and will be verified by the public key PK:

[0135] Verify(PK,H(D),Sig)=True or False,

[0136] Successful verification indicates that the integrity and legitimacy of the data are guaranteed.

[0137] In step S1, the real-time correlation formula between environmental data and RFID tag information is used to integrate multi-dimensional data to form a comprehensive evaluation index for the status of archives. Through the index, the system can monitor the status of archives in real time, organically combine environmental changes and RFID information, and fully grasp the status of archives;

[0138] To further ensure the security and reliability of data, the system uses digital signature technology to encrypt the generated comprehensive index of archive status S(t). The generation and verification process of digital signatures ensures the integrity and non-repudiation of data. Any tampering behavior is detected in time through the verification process. In addition, blockchain technology stores signatures and data to ensure the transparency and security of data and prevent unauthorized access and tampering.

[0139] By integrating algorithms, the system can provide accurate and real-time file status monitoring while ensuring data security, providing a solid technical foundation for intelligent file cabinet management methods.

[0140] In step S2, the adaptive signal enhancement mechanism and the sensor data fusion and tuning formula are used to optimize the power and frequency adjustment of the RFID reader to adapt to different environments, and dynamically adjust in combination with sensor data;

[0141] Adaptive signal enhancement mechanism formula:

[0142] P(t) is the output power of the RFID reader at time t; f(t) is the operating frequency of the RFID reader at time t; E(t) is the environmental data vector collected at time t; R(t) is the RFID signal strength at time t; γ(t) is the environmental impact factor at time t;

[0143] Calculation of environmental impact factors: Calculate the environmental impact factor γ(t) based on the environmental data E(t):

[0144]

[0145] Among them, α i is the weight of each environmental parameter, and m is the number of dimensions of environmental data;

[0146] Power adjustment formula: Adjust the output power P(t) of the RFID reader according to the size of γ(t):

[0147] P(t+1)=P(t)·(1+k1·γ(t)),

[0148] Wherein, k1 is the adjustment coefficient, which is used to control the adjustment amplitude of the power, and P(t+1) is the output power at time t+1;

[0149] Frequency adjustment formula: Adjust the operating frequency f(t) according to the signal strength R(t) and the environmental impact factor γ(t):

[0150]

[0151] Wherein, k2 is the frequency adjustment coefficient.

[0152] In step S2, the performance optimization of the RFID reader is achieved through the adaptive signal enhancement mechanism and the fusion and tuning formula of sensor data. First, the system calculates the environmental impact factor γ(t) by weighted processing of environmental data. The factor reflects the potential impact of the environment on the RFID signal. Second, the system dynamically adjusts the output power and operating frequency of the RFID reader according to the environmental impact factor and real-time signal strength.

[0153] The core of the adaptive signal enhancement mechanism is to automatically optimize the parameter settings of the RFID reader by sensing environmental changes in real time. The dynamic adjustment mechanism can maintain signal stability and reading accuracy in various complex environments to avoid signal loss or misreading. At the same time, the fusion and tuning of sensor data ensure the adaptability and intelligence of the RFID system, so that the system can flexibly respond to dynamic changes in the environment and ensure the efficiency and accuracy of file management.

[0154] Through the coordination of algorithm formulas, the system can realize intelligent adjustment of RFID readers to ensure their optimal performance under different environmental conditions, improve the application effect of RFID technology in archive management, and enhance the overall stability and reliability of the system.

[0155] Sensor data fusion and dynamic tuning formula To ensure the best performance of RFID readers in complex environments, the system fuses sensor data and performs dynamic tuning. The algorithm formula is:

[0156] S(t) is the weighted fusion result of sensor data; θ(t) is the RFID performance tuning parameter set at time t; L(θ(t)) is the performance loss function;

[0157] Sensor data fusion formula: Combining the data from different sensors, we get the fused environmental state index S(t):

[0158]

[0159] Among them, w j is the weight of each sensor data, D j (t) is the data of sensor j at time t, n is the number of sensors;

[0160] Dynamic tuning formula: Use the gradient descent method to optimize the tuning parameters θ(t) of the RFID system:

[0161]

[0162] Where η is the learning rate, is the gradient of the loss function L with respect to the parameter θ(t).

[0163] In step S2, sensor data fusion and dynamic tuning formula work together to optimize the performance of the RFID reader. First, the system integrates the data of multiple sensors through the sensor data fusion formula to form a comprehensive environmental state index S(t). The index fully reflects the state of the environment and provides a key reference for the performance tuning of the RFID reader.

[0164] On this basis, the system uses a dynamic tuning formula to continuously adjust the tuning parameters θ(t) of the RFID reader through the gradient descent method. By minimizing the performance loss function L(θ(t)), the system ensures the best performance of the RFID reader under different environmental conditions. The process is adaptive and dynamic, and can respond to environmental changes in a timely manner, adjust the output power, operating frequency and other parameters of the reader, and avoid signal interference, signal weakening and other problems.

[0165] The core of the sensor data fusion and dynamic tuning mechanism lies in the combination of environmental perception and system tuning, so that the RFID reader can intelligently adapt to complex and changing environmental conditions. Through the fusion and tuning of real-time data, the system can maintain the efficient operation of the RFID reader, improve the accuracy and stability of signal reading, and thus improve the efficiency and reliability of overall file management, ensuring that the system can continue to provide optimal performance in various complex environments and give full play to the role of RFID technology in intelligent file management.

[0166] In step S4, the archive query, positioning, status verification and blockchain recording formulas are used to optimize the accuracy of archive query, positioning accuracy, status verification and blockchain recording security;

[0167] File query and positioning algorithm formula: File query and positioning require comprehensive analysis based on RFID tag data, sensor data and camera image information:

[0168] Q(t) is the archive query request at time t, R(t) is the response data read by the RFID tag, C(t) is the image data obtained by the camera, and L(t) is the archive location information finally located by the system;

[0169] File location formula: File location is performed by integrating RFID data and image data:

[0170] L(t)=w1·R(t)+w2·Image_Process(C(t)),

[0171] Among them, w1 and w2 are the weight coefficients of RFID data and image data,

[0172] Image_Process(C(t)) represents the extraction of position information after image data processing.

[0173] Status verification algorithm formula: After locating the file, the system needs to verify the status of the file, whether it is complete and in the correct location:

[0174] S expected is the expected status of the archive recorded in the system, S actual (t) is the actual status of the archive acquired through sensor and image data at time t;

[0175] Status Verification Formula: Profile status verification is performed by comparing the expected status with the actual status:

[0176] ΔS(t)=S expected -S actual (t),

[0177] If ΔS(t) is less than the preset tolerance ∈, the state verification is considered to have passed:

[0178]

[0179] In step S4, archive query, location, status verification and data blockchain recording formula work together to optimize archive management and ensure security;

[0180] First, the archive query and positioning algorithm accurately locates the location of archives by comprehensively analyzing RFID tag data and camera image information. By integrating different data sources, the system can overcome the limitations of a single data source and achieve high-precision archive positioning, even in complex environments.

[0181] Next, the status verification algorithm further ensures that the files are located in the correct place and in the expected state. By comparing the difference between the actual state and the expected state, the system can promptly detect abnormalities in the files and ensure the accuracy and reliability of file management.

[0182] Finally, the blockchain recording and verification algorithm securely records and verifies the process of archive query, location and status verification. Through digital signatures and blockchain technology, the system ensures the immutability and traceability of archive operations, thereby improving the security and transparency of the management system.

[0183] The combination of algorithmic formulas makes the archive management system more accurate and reliable in terms of query, positioning and status verification. At the same time, the application of blockchain technology further ensures the security of data and the legality of operations, and builds an intelligent, efficient and secure archive management system. Its optimization measures improve the performance of the system and provide all-round security protection for archive management.

[0184] In step S5, full disk scanning, camera image verification, anomaly detection and big data analysis achieve comprehensive monitoring of file status and detection of anomalies through the fusion and intelligent analysis of multi-source data;

[0185] Anomaly detection identifies abnormal status of files by performing big data analysis on full disk scan and image verification data:

[0186] D(t) is the set of all archive status data scanned at time t, including S i (t) and image verification results, μ D is the expected value of the data set D(t), σ D is the standard deviation of the data set D(t);

[0187] Anomaly detection formula: Use the z-score method to detect anomalies in the status data of each file:

[0188] Z i(t) is the z-score of the i-th file status, and the z-score calculation formula is:

[0189]

[0190] If | Z i (t)|>Z threshold , then the file status is considered abnormal:

[0191]

[0192] In step S5, through full disk scanning, camera image verification, anomaly detection and big data analysis, the system achieves comprehensive monitoring of file status and intelligent detection of abnormal situations;

[0193] The z-score anomaly detection method is the core of the entire anomaly detection process. First, the system obtains the archive status data set D(t) at time t through full disk scanning and image verification, and calculates the mean μ of the data set. D and standard deviation σ D , and then calculate the Z of each archive state i (t) The system can quantify the degree of difference between the status of each file and the overall status in a standardized manner. Its standardized method effectively filters out the interference of environmental factors and focuses on truly abnormal files;

[0194] On this basis, the anomaly detection formula determines whether the file status is abnormal according to the size of the z-score. If the z-score value of a file exceeds the preset threshold, the system will mark it as an abnormal file. This process enables the system to identify and respond to abnormal situations in file management in a timely and accurate manner.

[0195] Through the coordination of algorithm formulas, the system realizes all-round intelligent monitoring of archive status. The introduction of the z-score method enables the system to detect abnormal archives in a standardized manner, effectively filter environmental noise, and accurately identify anomalies. The anomaly detection formula further strengthens the system's anomaly identification and processing capabilities, ensuring that various anomalies in the archive management process can be discovered and measures taken in a timely manner. Ultimately, with the support of big data analysis, the system improves the security, reliability, and intelligence level of archive management.

[0196] In step S6, abnormal operation identification, real-time alarm triggering, user operation permission management, and blockchain data auditing and verification are performed. The abnormal operation identification detects abnormal behavior by analyzing changes in user operation patterns and environmental data:

[0197] U(t) is the user's operation data at time t, H(t) is the environmental data change vector at time t, O(t) is the comprehensive operation behavior vector, which represents the combined information of operation and environment at time t, and θ is the standard parameter set of the operation mode;

[0198] Operation behavior vector formula: Combine user operation data U(t) with environmental data change H(t) to form an operation behavior vector:

[0199] O(t)=U(t)+λ·H(t),

[0200] Among them, λ is the weight coefficient;

[0201] Anomaly identification formula: Assumption O expected,i The expected value of the behavior vector for normal operation is used to determine whether the operation is abnormal using the distance measurement method:

[0202]

[0203] If Dist(O(t),O expected ) exceeds the set threshold ∈, it is judged as an abnormal operation:

[0204]

[0205] Real-time alarm triggering algorithm formula: If the system detects abnormal operation, the alarm should be triggered in time:

[0206] A(t) is the anomaly detection result at time t, and τ(t) is the time when the alarm is triggered at time t;

[0207] Real-time alarm formula: If the anomaly detection result is true, the system triggers an alarm:

[0208]

[0209] The operation behavior vector formula combines user operation data with environmental changes to form a comprehensive operation behavior vector O(t), which enables the system to take into account the impact of environmental factors when detecting abnormal operations. Its comprehensive perspective improves the system's sensitivity to abnormal operations and enables it to more accurately reflect the actual situation of user behavior.

[0210] Next, the anomaly identification formula accurately identifies abnormal operations by calculating the distance between the operation behavior vector and the normal operation mode. If the operation behavior is too far away from the expected behavior, the system will mark the operation as abnormal. Based on the distance measurement method, the system can effectively identify potential anomalies in user behavior, especially in the case of multi-variable and multi-dimensional data, and can achieve accurate anomaly detection;

[0211] Once an abnormal operation is detected, the real-time alarm formula ensures that the system can immediately trigger an alarm, notify relevant personnel or activate the automatic protection mechanism. The immediate response mechanism significantly improves the security of the system, enabling it to respond to abnormal operations as soon as possible and reduce potential security risks.

[0212] In step S7, the RFID reader signal adaptive adjustment algorithm formula automatically optimizes its working parameters according to the changes in environmental conditions for the prediction of equipment failure and early detection of tag failure.

[0213] P(t) is the output power of the RFID reader at time t, f(t) is the operating frequency of the RFID reader at time t, and E(t) is the environmental impact factor at time t;

[0214] Output power adjustment formula: Adjust the output power of the RFID reader according to the environmental impact factor E(t):

[0215] P(t+1)=P(t)·(1+k p E(t)),

[0216] Among them, k p is the power adjustment coefficient, which controls the sensitivity of power adjustment;

[0217] Working frequency adjustment formula: Adjust the working frequency of the RFID reader according to environmental changes E(t) and signal quality Q(t):

[0218]

[0219] Among them, k f is the frequency adjustment coefficient, ∈ is a small constant to avoid the denominator being zero, and Q(t) represents the signal quality indicator.

[0220] In step S7, the adaptive adjustment of RFID reader signals, prediction of equipment failures, and early detection of tag failures are designed to automatically optimize the operating parameters of the RFID reader through real-time monitoring and analysis of environmental conditions to ensure the stable operation of the system in complex environments;

[0221] The output power adjustment formula dynamically adjusts the output power of the RFID reader by real-time monitoring of environmental influencing factors to adapt to environmental changes. If environmental interference increases, the system will automatically increase the output power to ensure the stability and reading accuracy of the RFID signal. The dynamic power adjustment mechanism ensures the reliability of the system under various environmental conditions.

[0222] The working frequency adjustment formula further enhances the adaptability of RFID signals. By combining signal quality indicators and environmental factors, the system can automatically adjust the working frequency of the reader to enhance the signal's anti-interference ability. The adaptive adjustment of the frequency ensures that the system can maintain a good working state when facing complex and changeable environmental interference.

[0223] The formula works together for the adaptive optimization of the RFID reader, enabling the system to automatically adjust operating parameters according to real-time environmental changes. Through dynamic adjustment of power and frequency, the system can maintain signal stability and reading accuracy in various complex environments, and can effectively reduce the risk of equipment failure and tag failure. Ultimately, the adaptive adjustment mechanism significantly improves the application effect of RFID technology in intelligent archive management and ensures the reliability and efficiency of the system.

[0224] In step S8, the multidimensional data collection and aggregation algorithm formula is used for the collection, analysis and optimization of the archive management strategy of multidimensional data and the automatic upgrade and expansion of the system:

[0225] D i (t) is the data vector collected by the i-th sensor or device at time t, W i is the weight of the i-th data source, M(t) is the multidimensional data aggregation result at time t;

[0226] Data aggregation formula: aggregate data from different sources according to weights;

[0227]

[0228] Where n is the number of data sources.

[0229] First, the data aggregation formula forms a comprehensive, multi-dimensional data view by weighted aggregation of inputs from different data sources. Data sources include environmental sensors, RFID readers, cameras, etc. The contribution of each data source in data integration is determined by the weight. By setting the weight reasonably, the system can highlight the impact of key data sources and enhance the pertinence and accuracy of data analysis.

[0230] Based on the results of multi-dimensional data aggregation, the system can conduct in-depth analysis to identify potential optimization opportunities and improvement directions. For example, the system can analyze aggregated data to identify weak links in archive management strategies and make optimization suggestions to improve management efficiency. At the same time, aggregated data provides a decision-making basis for automatic upgrades and expansions of the system, so that the system can make adaptive adjustments based on actual needs.

[0231] Through the collection and weighted aggregation of multi-dimensional data, the system can fully understand the various environmental factors and operational data involved in the archive management process, and provide strong data support for strategy optimization and system expansion. The data aggregation formula integrates multi-source data and improves the accuracy and pertinence of data analysis through weight setting. Ultimately, the multi-dimensional data collection and analysis method helps the system to achieve more intelligent archive management, support the system's automated upgrades and flexible expansion, and improve the overall management effect and system adaptability.

[0232] In step S8, the system self-learning and model self-adaptation formulas are used to cope with the ever-changing needs. The system must have self-learning and self-adaptation capabilities:

[0233] θ(t) is the set of model parameters at time t, and L(θ(t)) is the loss function of the model, which measures the performance of the model;

[0234] Self-learning update formula: Use adaptive learning rate to perform self-learning update of model parameters:

[0235]

[0236] Among them, the learning rate η(t) is dynamically adjusted according to the model performance to accelerate the learning process:

[0237] η(t+1)=η(t)·(1+κ·ΔL(t)),

[0238] Among them, κ is the learning rate adjustment coefficient, and ΔL(t) is the rate of change of the loss function.

[0239] In step S8, the system self-learning and model adaptation formula provides key support for the system to cope with changing needs. Through the formula, the system can continuously adjust and optimize its own model parameters to adapt to new environments and new needs, while maintaining efficient and accurate operation;

[0240] The self-learning update formula optimizes the model parameter set through the gradient descent method to ensure that the system can continuously improve the performance of the model in a dynamic environment. The system uses the gradient information of the loss function to identify the deficiencies of the model and make targeted parameter adjustments to enable the model to continuously develop in a better direction.

[0241] To enhance the intelligence of the learning process, the adaptive learning rate adjustment formula dynamically adjusts the learning rate, so that the system can adjust its learning rate according to the actual learning effect. The mechanism allows the system to speed up the learning pace when the effect is significant and quickly adapt to new needs. When the learning effect is not obvious, the learning rate is reduced to ensure stability and avoid instability caused by too fast adjustment.

[0242] Through self-learning and adaptive mechanisms, the system has the ability to continuously optimize in complex and changing environments. The formulas work together to enable the system to dynamically adjust model parameters and learning rates, ensuring that the system can quickly adapt and make efficient and accurate decisions when responding to new demands and new challenges. Ultimately, the self-learning and adaptive capabilities significantly improve the intelligence level of the system, enabling it to maintain excellent performance and high adaptability in long-term operation.

[0243] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A smart filing cabinet management method based on RFID technology, characterized in that: The method comprises the following steps: S1, entering the tag information and the file information into the system, and encrypting and storing them through the blockchain technology. At the same time, arranging sensors in the file cabinet to collect environmental data and associate them with the RFID tag information in real time; S2. Install RFID readers, sensors and cameras, and debug their working status to collect data in various environments. In addition, an adaptive signal enhancement mechanism is introduced to adjust the power and frequency of the RFID readers according to the real-time environment. S3. Put the file into the smart filing cabinet. The system automatically reads the RFID tag and records the storage location of the file. It also collects and records environmental data. The information is encrypted and stored in the blockchain. S4. Query through the archive management system. The system can accurately locate and verify the archive status based on RFID and sensor data combined with camera images. When taking out or returning archives, the system automatically updates the archive status and records the operation information in combination with blockchain. S5. Regularly trigger RFID readers and sensors to scan the entire disk. At the same time, use camera images to verify the status of the file. The system detects anomalies through big data analysis and machine learning algorithms, generates inventory reports, and marks abnormal situations. S6. By analyzing the file usage patterns and environmental data, the system automatically identifies abnormal operations and issues real-time alarms to prevent data loss or file damage. At the same time, the user's operation permissions are strictly managed through blockchain technology, and recorded in the blockchain ledger, and blockchain data is audited regularly; S7. The system automatically adjusts the signal parameters of the RFID reader according to environmental changes, and analyzes the operating status of the equipment through machine learning algorithms, predicts equipment failure or tag failure, and performs maintenance or replacement in advance; S8. Utilize the collected multi-dimensional data to optimize the archive management strategy through big data analysis, adjust the storage layout, and regularly upgrade the system software and expand the hardware according to the development of new technologies and changes in management needs.

2. According to the RFID technology-based intelligent filing cabinet management method of claim 1, it is characterized in that: In the step S1, in the real-time association and encrypted storage of the environmental data and the RFID tag information, the real-time association formula of the environmental data and the RFID tag information is as follows: E(t) represents the environmental data vector collected at time t; R(t) represents the RFID tag information vector collected at time t; α i is the weight coefficient of different environmental parameters; β j is the weight coefficient of different RFID data parameters; S(t) represents the comprehensive index of archive status calculated at time t; Among them, m is the dimension number of environmental data, and n is the dimension number of RFID data; To ensure the integrity and non-repudiation of data, digital signature technology is used, and its algorithm formula is: SK is the private key, used for signing; PK is the public key, used for verifying the signature; Sig is the signature result; The digital signature generation process is expressed as: Sig=Sign(SK,H(D)), The transaction data in the blockchain will include Sig and H(D) and will be verified by the public key PK: Verify(PK,H(D),Sig)=True or False, Successful verification indicates that the integrity and legitimacy of the data are guaranteed.

3. The intelligent filing cabinet management method based on RFID technology according to claim 1 is characterized in that: In the step S2, the adaptive signal enhancement mechanism and the sensor data fusion and tuning formula are used to optimize the power and frequency adjustment of the RFID reader to adapt to different environments, and dynamically adjust in combination with the sensor data; The adaptive signal enhancement mechanism formula: P(t) is the output power of the RFID reader at time t; f(t) is the operating frequency of the RFID reader at time t; E(t) is the environmental data vector collected at time t; R(t) is the RFID signal strength at time t; γ(t) is the environmental impact factor at time t; Calculation of environmental impact factors: Calculate the environmental impact factor γ(t) based on the environmental data E(t): Among them, α i is the weight of each environmental parameter, and m is the number of dimensions of environmental data; Power adjustment formula: Adjust the output power P(t) of the RFID reader according to the size of γ(t): P(t+1)=P(t)·(1+k1·γ(t)), Wherein, k1 is the adjustment coefficient, which is used to control the adjustment amplitude of the power, and P(t+1) is the output power at time t+1; Frequency adjustment formula: Adjust the operating frequency f(t) according to the signal strength R(t) and the environmental impact factor γ(t): Wherein, k2 is the frequency adjustment coefficient.

4. The intelligent filing cabinet management method based on RFID technology according to claim 3 is characterized in that: The sensor data fusion and dynamic tuning formula is to ensure the best performance of the RFID reader in a complex environment. The system fuses the sensor data and performs dynamic tuning. The algorithm formula is: S(t) is the weighted fusion result of sensor data; θ(t) is the RFID performance tuning parameter set at time t; L(θ(t)) is the performance loss function; Sensor data fusion formula: Combining the data from different sensors, we get the fused environmental state index S(t): Among them, w j is the weight of each sensor data, D j (t) is the data of sensor j at time t, n is the number of sensors; Dynamic tuning formula: Use the gradient descent method to optimize the tuning parameters θ(t) of the RFID system: Where η is the learning rate, is the gradient of the loss function L with respect to the parameter θ(t).

5. The intelligent filing cabinet management method based on RFID technology according to claim 1 is characterized in that: In the step S4, the blockchain recording formulas for archive query, positioning, status verification and data are used to optimize the accuracy of archive query, the accuracy of positioning, the verification of status and the security of blockchain recording; The file query and positioning algorithm formula: File query and positioning require comprehensive analysis based on RFID tag data, sensor data and camera image information: Q(t) is the archive query request at time t, R(t) is the response data read by the RFID tag, C(t) is the image data obtained by the camera, and L(t) is the archive location information finally located by the system; File location formula: File location is performed by integrating RFID data and image data: L(t)=w1·R(t)+w2·Image_Process(C(t)), Among them, w1 and w2 are the weight coefficients of RFID data and image data, Image_Process(C(t)) represents the extraction of position information after image data processing. Status verification algorithm formula: After locating the file, the system needs to verify the status of the file, whether it is complete and in the correct location: S expected is the expected status of the archive recorded in the system, S actual (t) is the actual status of the archive acquired through sensor and image data at time t; Status Verification Formula: Profile status verification is performed by comparing the expected status with the actual status: ΔS(t)=S expected -S actual (t), If ΔS(t) is less than the preset tolerance ∈, the state verification is considered to have passed:

6. The intelligent filing cabinet management method based on RFID technology according to claim 1 is characterized in that: In the S5 step, full disk scanning, camera image verification, anomaly detection and big data analysis realize comprehensive monitoring of the archive status and detection of abnormal situations through the fusion and intelligent analysis of multi-source data; Anomaly detection identifies abnormal status of files by performing big data analysis on full disk scan and image verification data: D(t) is the set of all archive status data scanned at time t, including S i (t) and image verification results, μ D is the expected value of the data set D(t), σ D is the standard deviation of the data set D(t); Anomaly detection formula: Use the z-score method to detect anomalies in the status data of each file: Z i (t) is the z-score of the i-th file status, and the z-score calculation formula is: If | Z i (t)|>Z threshold , then the file status is considered abnormal:

7. The intelligent filing cabinet management method based on RFID technology according to claim 1 is characterized in that: In the step S6, abnormal operation identification, real-time alarm triggering, user operation authority management and blockchain data auditing and verification are performed. The abnormal operation identification detects abnormal behavior by analyzing the changes in user operation patterns and environmental data: U(t) is the user's operation data at time t, H(t) is the environmental data change vector at time t, O(t) is the comprehensive operation behavior vector, which represents the combined information of operation and environment at time t, and θ is the standard parameter set of the operation mode; Operation behavior vector formula: Combine user operation data U(t) with environmental data change H(t) to form an operation behavior vector: O(t)=U(t)+λ·H(t), Among them, λ is the weight coefficient; Anomaly identification formula: Assumption O expected,i The expected value of the behavior vector for normal operation is used to determine whether the operation is abnormal using the distance measurement method: If Dist(O(t),O expected ) exceeds the set threshold ∈, it is judged as an abnormal operation: Real-time alarm triggering algorithm formula: If the system detects abnormal operation, the alarm should be triggered in time: A(t) is the anomaly detection result at time t, and τ(t) is the time when the alarm is triggered at time t; Real-time alarm formula: If the anomaly detection result is true, the system triggers an alarm:

8. The intelligent filing cabinet management method based on RFID technology according to claim 1 is characterized in that: In the step S7, the RFID reader signal is adaptively adjusted, equipment failure is predicted, and tag failure is detected in advance. The RFID reader signal adaptive adjustment algorithm automatically optimizes its operating parameters according to changes in environmental conditions: P(t) is the output power of the RFID reader at time t, f(t) is the operating frequency of the RFID reader at time t, and E(t) is the environmental impact factor at time t; Output power adjustment formula: Adjust the output power of the RFID reader according to the environmental impact factor E(t): P(t+1)=P(t)·(1+k p ·E(t)), Among them, k p is the power adjustment coefficient, which controls the sensitivity of power adjustment; Working frequency adjustment formula: Adjust the working frequency of the RFID reader according to environmental changes E(t) and signal quality Q(t): Among them, k f is the frequency adjustment coefficient, ∈ is a small constant to avoid the denominator being zero, and Q(t) represents the signal quality indicator.

9. The intelligent filing cabinet management method based on RFID technology according to claim 1, characterized in that: In the step S8, the multidimensional data collection and aggregation algorithm formula is used for the collection and analysis of multidimensional data, the optimization of archive management strategy and the automatic upgrade and expansion of the system: D i (t) is the data vector collected by the i-th sensor or device at time t, W i is the weight of the i-th data source, M(t) is the multidimensional data aggregation result at time t; Data aggregation formula: aggregate data from different sources according to weights; Where n is the number of data sources.

10. The intelligent filing cabinet management method based on RFID technology according to claim 1, characterized in that: In the step S8, the system self-learning and model self-adaptation formula is to cope with the ever-changing needs. The system must have self-learning and self-adaptation capabilities: θ(t) is the set of model parameters at time t, and L(θ(t)) is the loss function of the model, which measures the performance of the model; Self-learning update formula: Use adaptive learning rate to perform self-learning update of model parameters: Among them, the learning rate η(t) is dynamically adjusted according to the model performance to accelerate the learning process: η(t+1)=η(t)·(1+κ·ΔL(t)), Among them, κ is the learning rate adjustment coefficient, and ΔL(t) is the rate of change of the loss function.

Images