Archival data access control method and system based on multi-level encryption
By combining multi-level encryption methods with correlation analysis of RSSI and acceleration data, the security blind spots of traditional solutions in complex environments are solved, achieving highly accurate behavior verification and autonomous security protection, thus improving the security of archival data.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WEINAN JINDUN ESCORT CO LTD
- Filing Date
- 2026-05-15
- Publication Date
- 2026-06-23
AI Technical Summary
Traditional data protection solutions are ill-suited to complex environments involving the handover of security personnel on foot, leading to accidental data locking and security blind spots. They fail to effectively distinguish between normal walking and abnormal movement, resulting in significant security blind spots and risks of delayed response.
By using a multi-level encryption method, an association weight matrix is constructed using the RSSI value and triaxial acceleration data received by the terminal. A bipartite graph maximum weight matching algorithm is used for behavior verification, and the trust score is dynamically updated to control the generation and destruction of the decryption key.
It improves the accuracy of behavior verification and the robustness of the system, realizes autonomous security protection in complex environments, reduces the false judgment rate, and enhances the operational experience of escorts.
Smart Images
Figure CN122268587A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data security management technology, specifically to a method and system for access control of archive data based on multi-level encryption. Background Technology
[0002] In financial escort operations, the security of archival data is paramount. However, traditional protection solutions rely excessively on near-field wireless communication (such as Bluetooth and Wi-Fi) between vehicles and handheld terminals for on-site monitoring, making them ill-suited for complex scenarios where escorts must leave the vehicle, carry the file boxes, and conduct handovers on foot. Existing judgment mechanisms are primarily based on a single signal strength index (RSSI) threshold. This static logic proves inadequate in urban high-rises, canyons, or indoor environments: strong multipath effects can cause drastic fluctuations in RSSI values, easily triggering terminal misjudgments. Terminals may mistake normal signal fluctuations for equipment movement away, leading to accidental data locking and severely disrupting normal operational processes.
[0003] Furthermore, single-dimensional signal monitoring cannot construct an effective behavioral feature map, resulting in the system's lack of ability to detect abnormal movement. Since relying solely on RSSI cannot distinguish whether the escort is in a standard weighted walking state or using tools such as skateboards or vehicles for rapid abnormal transfer, attackers can carry out "stealthy" hijacking while maintaining signal connectivity. In addition, in densely populated high-rise areas or indoor environments, GPS positioning, which serves as an auxiliary verification, often fails due to signal loss or drift. This makes traditional solutions have significant security blind spots and response delay risks when facing well-planned man-in-the-middle attacks or rapid transfer scenarios. Summary of the Invention
[0004] In view of the above, it is necessary to provide a method and system for access control of archive data based on multi-level encryption to solve the above problems.
[0005] The first aspect of this application provides a method for access control of archive data based on multi-level encryption, applied to a portable terminal storing encrypted archive data, the method comprising: Based on a preset number of consecutive RSSI values received by the terminal from a preset beacon device, a signal strength benchmark is determined. At the same time, based on the triaxial acceleration data collected by the terminal, an environmental noise vibration benchmark is obtained. One-dimensional synthetic modulus of triaxial acceleration data after removing gravity component is obtained. Foot impact judgment threshold is dynamically set based on the environmental noise vibration benchmark. The fluctuation characteristics of the modulus in the preset sliding observation window are compared with the foot impact threshold to generate foot vibration event. The continuous decrease change characteristics of RSSI value after detrending processing are analyzed to generate signal drop event. The differences between the corresponding attributes of foot vibration events and signal drop events are analyzed, an association weight matrix is constructed, and a bipartite graph maximum weight matching algorithm is used to search the association weight matrix to obtain matching paths and perform effective matching filtering. A non-matching penalty term is introduced, and a walking signal consistency score is calculated. The trust score is dynamically updated based on the walking signal consistency score, and the generation and destruction of the decryption key corresponding to the encrypted file data are controlled when the trust score meets the threshold condition; and the destruction of the decryption key is triggered when the real-time RSSI value is continuously higher than the return threshold set based on the signal strength benchmark.
[0006] Preferably, the determination of the signal strength reference specifically includes: If the variance of the fluctuation of L consecutive RSSI values is lower than the preset stability threshold and the mean is higher than the effective connection threshold, then the mean of the L RSSI values is used as the signal strength benchmark, where L is the preset number.
[0007] Preferably, the process of obtaining the environmental noise floor vibration reference is as follows: The root mean square value of the filtered acceleration data after a preset duration is used as the initial value of the environmental noise floor vibration reference, denoted as... The preset minimum effective vibration threshold is denoted as... ; like ,Will As the final environmental noise floor vibration benchmark; if The pre-calibrated default noise floor parameter is used as the final vibration amplitude benchmark.
[0008] Preferably, the specific process of setting the foot impact judgment threshold is as follows: The product of the preset signal-to-noise ratio coefficient and the environmental background noise vibration benchmark, plus the preset safety margin value, is used as the threshold for judging foot impact.
[0009] Preferably, the specific process of generating the footstep vibration event is as follows: Based on all the magnitude values obtained in the sliding observation window, the peak corresponding to the maximum value that is greater than the foot impact judgment threshold is defined as a foot vibration event. The foot vibration event includes two attributes: the time of occurrence and the impact amplitude.
[0010] Preferably, the signal drop event is specifically: Based on all adjacent RSSI values after the trend has ended in the sliding observation window, the signal rate of change is calculated. When the signal rate of change is less than the preset signal drop threshold, and the time the signal maintains a downward trend exceeds the minimum drop duration, and the magnitude of the drop exceeds the preset minimum effective drop value, the corresponding sliding observation window is taken as the drop interval and defined as generating a signal drop event. The generated signal drop event includes two attributes: the time of the drop and the magnitude of the drop; the time of the drop is obtained by the midpoint of the drop interval; the magnitude of the drop is obtained by the difference between the signal value at the starting point and the signal value at the lowest point.
[0011] Preferably, the construction of the correlation weight matrix specifically involves: The first in the correlation weight matrix Line number The specific elements of the column are: the first... The first footstep vibration event and the first Correlation weights among signal drop events The specific formula is as follows: in, This represents the time interval between the i-th foot vibration event and the j-th signal drop event; The default time tolerance is indicated by exp(); exp() represents an exponential function with the natural constant as the base.
[0012] Preferably, the specific process for calculating the consistency score of the walking signal is as follows: The number of valid matches with a matching weight greater than a preset matching threshold in the matching path is denoted as . ; The maximum number of rows and columns in the correlation weight matrix is used as the maximum total number of events within the current window, denoted as . ; Walking signal consistency score The calculation formula is as follows: in, This indicates the preset penalty coefficient; The index set represents the valid matching pairs; s represents the index of the valid matching pairs. This represents the matching weight of the s-th valid matching pair.
[0013] Preferably, the process of dynamically updating the trust score based on the walking signal consistency score specifically includes: If the walking signal consistency score If the score exceeds the preset compliance threshold, a bonus will be awarded: ; If the walking signal consistency score If the score is below the preset abnormal threshold, a point deduction will be performed. ; Otherwise, the trust score will decrease by a preset decay rate; in, , Let these represent the trust integrals of the k-th and (k-1)-th sliding observation windows, respectively. The preset maximum score. The preset step size for adding points is defined by k, which represents the discrete time step; min() represents the function to be minimized. The preset deduction step size is greater than the addition step size; max() represents the maximization function.
[0014] Secondly, embodiments of this application also provide a file data access control system based on multi-level encryption, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.
[0015] This application has at least the following beneficial effects: First, this application improves the accuracy of behavior verification in complex environments: by introducing a bipartite graph maximum weight matching algorithm, this application can effectively handle the time misalignment and quantity mismatch between footstep vibration and signal drop. Even in scenarios with multipath interference, it can accurately identify real behaviors that conform to the characteristics of carrying weight based on statistical regularities, filtering out abnormal movement patterns such as skateboard transport and vehicle pick-up.
[0016] Secondly, the robustness and availability of the system are enhanced: the access control mechanism based on the trust points pool allows the system to tolerate short-term signal interruptions or misjudgments while ensuring security, avoiding frequent lockouts caused by single data point anomalies, and significantly improving the operational experience of the escort personnel.
[0017] Finally, autonomous security protection without relying on external base stations is achieved: by using vehicle engine vibration and in-vehicle signal strength as dynamic benchmarks, this application does not rely on GPS or external base stations and can work independently in signal-free areas such as basements and shielded equipment rooms using only the terminal's own sensors, filling the gap in existing offline escort security technology.
[0018] In summary, this application effectively solves the problems of difficult behavior verification and high false judgment rate in offline escort scenarios through deep correlation analysis of multimodal features, and significantly improves the security of archive data. Attached Figure Description
[0019] Figure 1 A flowchart illustrating the steps of a file data access control method based on multi-level encryption, provided in one embodiment of this application. Detailed Implementation
[0020] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.
[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this application's specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.
[0022] It should also be noted that the terms "first" and "second" in this application and its accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the method. Without departing from the scope of protection of this application, the execution order of multiple steps can be interchanged, and some steps can also be deleted.
[0023] The following description, in conjunction with the accompanying drawings, details the specific scheme of the file data access control method and system based on multi-level encryption provided in this application.
[0024] Please see Figure 1 This document illustrates a flowchart of a multi-level encryption-based archive data access control method according to an embodiment of this application. The scenario considered is a portable terminal held by a security guard, which stores encrypted archive data. Offline dynamic authorization is achieved by scanning the vehicle's OBU signal based on the following steps: The first step is to determine the signal strength benchmark based on a preset number of consecutive RSSI values received by the terminal from the preset beacon device, and at the same time, to obtain the environmental noise vibration benchmark based on the triaxial acceleration data collected by the terminal.
[0025] Step 101: Acquire signal strength benchmark.
[0026] First, the Bluetooth Low Energy (BLE) module of the terminal containing the encrypted file is activated to continuously scan for broadcast signals emitted by the onboard unit (OBU). To eliminate interference from transient signal fluctuations, this application maintains a length of... (In this embodiment, the value is taken as) ) signal sampling queue .
[0027] Whenever a terminal captures a broadcast packet, it pushes its Received Signal Strength Indicator (RSSI) value into a queue. Real-time monitoring queue. The values in the range are used to determine the current environment as in-vehicle. When the variance of L consecutive RSSI values is lower than a preset stability threshold (5 in this embodiment) and the mean is higher than the effective connection threshold (-45dBm in this embodiment, representing the device being adjacent to the beacon and within a metal enclosed space), the current environment is considered to meet the characteristics of an in-vehicle environment. At this point, an arithmetic mean calculation is performed, and the average value of all RSSI values in the signal sampling queue is used as the signal strength benchmark, denoted as [missing value]. .
[0028] Step 102: Collect environmental noise and vibration baselines and perform engine shutdown as a fallback.
[0029] In lock Within the same time window, the inertial measurement unit (IMU) is activated to collect triaxial acceleration data at a fixed sampling rate (50Hz in this embodiment). First, a band-stop filter is applied to the raw acceleration data (the stopband range is set to 0.5Hz to 3.0Hz in this embodiment) to filter out low-frequency components such as personnel micro-movements or vehicle shaking, retaining only the high-frequency vibration components unique to the engine.
[0030] Then, select a period of time. (In this embodiment, the interval is 2 seconds; the implementer can adjust this as needed.) The filtered acceleration data is used to calculate its root mean square (RMS) value, which serves as the initial value for the environmental noise vibration baseline. To prevent the reference value from becoming too low (proximity sensor noise floor) due to the vehicle engine stalling, a minimum effective vibration threshold is introduced. (Set to 0.02g).
[0031] If the actual collected value ,Will As the ultimate environmental noise and vibration benchmark If the actual collected value This application will use pre-calibrated default noise floor parameters The final vibration amplitude benchmark (0.1g in this embodiment) is used as the benchmark, and the terminal is marked to enter the high-sensitivity monitoring mode.
[0032] Step 103: Dynamic benchmark update mechanism.
[0033] Considering that vehicle movement or environmental changes may cause baseline drift, the terminal continuously monitors the real-time RSSI value during operation. If the monitored real-time RSSI remains above the return-to-base threshold... ;in, This is a preset value used to represent the tolerance range; in this embodiment, it is set to 3 dBm; and the duration exceeds... (In this embodiment, it is set to 10 seconds), and it is determined that the terminal device has returned to the safe area inside the vehicle. At this time, the terminal automatically triggers steps 101 and 102 to re-collect and update data. and In order to adapt to the new environmental conditions.
[0034] The second step: Time alignment and trend separation processing are performed on the RSSI and triaxial acceleration data.
[0035] This step mainly involves performing time alignment, trend separation, and normalization on the raw data stream from the sensors to address the issues of inconsistent sampling rates and dimensions among different sensors, thus providing standardized data input for subsequent correlation analysis.
[0036] Step 201: Time alignment of multi-source data.
[0037] Since the broadcast interval of the Bluetooth module (100ms, or 10Hz in this embodiment) is usually greater than the sampling interval of the IMU (20ms, or 50Hz in this embodiment), direct comparison will lead to data misalignment. The terminal uses the high-frequency sampling timestamp of the IMU as a reference to perform linear interpolation on the low-frequency RSSI data.
[0038] Specifically, for any two adjacent original RSSI sampling points and ,like Take directly As a result, skip the interpolation calculation; otherwise, at time... ( The RSSI value of ) is calculated using the following formula: This step generates a dense RSSI sequence that is fully synchronized with the acceleration data, ensuring that millisecond-level signal drop details can be captured subsequently.
[0039] Step 202: Signal detrending processing.
[0040] As the escort personnel move away from the vehicle, the RSSI (Residual RSSI) includes not only high-frequency fluctuations caused by torso occlusion but also low-frequency attenuation trends due to increased distance. To extract pure occlusion features, this application employs a moving average method to separate the trend term.
[0041] A long window is preset (in this embodiment, the time length corresponding to the long window is 10 ... (The duration is [number of seconds], which the implementer can adjust according to the actual situation). Calculate the moving average of the RSSI data after interpolation within this long window. Subsequently, the detrended signal fluctuation components are calculated. : It should be noted that when the data length is insufficient... In this case, the arithmetic mean of the currently collected data is used instead of the moving average. This signal fluctuation component eliminates the overall attenuation caused by distance, retaining only the instantaneous fluctuation characteristics caused by occlusion or multipath effects, providing a cleaner input for subsequent drop detection.
[0042] The third step is to obtain the one-dimensional composite modulus of the triaxial acceleration data after removing the gravity component, dynamically set the foot impact judgment threshold based on the environmental noise vibration benchmark, compare the fluctuation characteristics of the modulus in the preset sliding observation window with the foot impact threshold, and generate a foot vibration event; analyze the continuous decrease characteristics of the RSSI value after detrending processing to generate a signal drop event.
[0043] The core task of this step is to transform continuous sensor waveforms into discrete physical events, namely foot vibration events and signal drop events. This discretization process can significantly reduce the computational complexity of subsequent correlation analysis and focus on the most behaviorally significant key moments.
[0044] Step 301: Foot vibration extraction based on dynamic threshold.
[0045] First, the triaxial acceleration data is processed by passing it through a high-pass filter (cutoff frequency 0.5Hz) to remove the gravitational acceleration component, and then its magnitude is calculated (i.e., The overall motion intensity is characterized by a vibration frequency (FQF). To accurately distinguish between vehicle engine vibration (high-frequency, low-amplitude) and human walking impact (low-frequency, high-amplitude), an environmental noise floor vibration benchmark is used. Set a dynamic detection threshold.
[0046] Specifically, obtain a duration of (In this embodiment, it is set) A sliding observation window of seconds (in seconds), with a step size of 1. (In this embodiment, the value is 0.5 seconds), and adjacent sliding observation windows contain an overlapping area of a preset time length. For the data within the current window, the following peak detection logic is executed: (1) Threshold calculation: Set the foot impact judgment threshold. This threshold is not a fixed value, but is dynamically adjusted based on engine baselines. ;in, The preset signal-to-noise ratio coefficient (in this embodiment) ), The preset safety margin is 0.1g in this embodiment. This formula ensures that the threshold can be adaptively adjusted under different vehicle models (with different vibration levels) and different operating conditions (such as idling or driving), so as not to miss faint footsteps or misdetect vehicle bumps.
[0047] (2) Peak search: Traverse the acceleration magnitude sequence within the window and find local maxima that satisfy the following two conditions: a) Amplitude condition: the magnitude at this point ; b) Extreme value condition: The magnitude of the sampling points before and after this point is less than that point.
[0048] (3) Event Generation: Each wave peak that meets the conditions is defined as an independent foot vibration event. Assume that a total of [number] wave peaks are detected within the window. These wave peaks generate a set of foot vibration events. Among them, the first One event It contains two attributes: the time of occurrence. and impact amplitude .
[0049] It should be noted that for the set of all impact amplitudes extracted by the sliding observation window, the max-min normalization method is performed within each sliding observation window to calculate its normalized value. When performing the max-min normalization method, to avoid the denominator being 0, a very small positive number needs to be added to the denominator; in this embodiment, the value is 0.01.
[0050] Step 302: Extraction of drop events based on detrending signals.
[0051] Parallel processing of detrended signal fluctuation components Since this component has eliminated the distance decay trend, its negative fluctuations mainly reflect instantaneous occlusion or multipath effects. To extract significant fall losses caused by torso occlusion, this application employs a gradient detection method: (1) Gradient calculation: For The first-order difference (i.e., the difference between adjacent points) of the sequence is calculated to obtain the rate of change of the signal. Because human occlusion usually causes a sharp drop in signal strength within a short period of time, the terminal focuses on segments with negative change rates and large absolute values.
[0052] (2) Drop detection: Set a signal drop detection threshold. (In this embodiment, it is -3dBm / 100ms). The terminal scans the sequence to find time points that meet the following conditions: a) Slope condition: the rate of change at that point ; b) Persistence condition: The signal maintains a downward trend for a period exceeding the duration of the lowest drop. (In this embodiment, the value is 50ms) to eliminate high-frequency noise interference.
[0053] (3) Amplitude quantization: For each detected drop interval, calculate its drop amplitude. This refers to the signal difference between the starting point and the lowest point of the interval. For the accuracy of subsequent correlation calculations, this application only retains... An event that exceeds the minimum effective drop value (2dBm in this embodiment).
[0054] (4) Event Generation: Define each falling interval that meets the conditions as an independent event. Assume that a total of [number] falls are detected within the window. This type of interval generates a set of signal drop events. Among them, the first One event It contains two attributes: the time when the fall occurred. (Take the midpoint of the interval) and the magnitude of the drop .
[0055] For the set of all signal drop amplitudes The normalized value is calculated using the minimum-maximum normalization method. It should be noted that, when performing the max-min normalization method, in order to avoid the denominator being 0, a very small positive number needs to be added to the denominator. In this embodiment, the value is 0.01.
[0056] This application successfully abstracts the originally chaotic continuous sensor data into two sets of discrete events with clear physical meaning. and This is used to subsequently calculate the causal association weights, thereby verifying the authenticity of the behavior.
[0057] The fourth step is to analyze the differences in the corresponding attributes of foot vibration events and signal drop events, construct an association weight matrix, use a bipartite graph maximum weight matching algorithm to search the association weight matrix, obtain matching paths and perform effective matching filtering, introduce a non-matching penalty term, and calculate the walking signal consistency score.
[0058] For foot vibration event set and signal drop event set This step addresses the issue that the two are not perfectly aligned in terms of quantity and time (i.e., there are instances of more steps than falls or more falls than steps), and uses a bipartite graph matching algorithm to find the optimal explanatory path between them, thereby quantifying the consistency of the overall behavior.
[0059] Step 401: Construct the time-amplitude correlation weight matrix.
[0060] To assess whether each footstep vibration is accompanied by a reasonable signal drop, it is first necessary to calculate the vibration of any footstep in the set. With any fall The probability magnitude between them. This application constructs a dimension of... Correlation weight matrix ,in, For the number of steps, This represents the number of drops.
[0061] For the nth matrix Line number The element of the column, i.e., the first A footstep vibration event With the Signal drop event Association weight between The calculation is performed using the Gaussian kernel function. This function maps the time difference to the amplitude difference. The probability values between them, specifically: in, : Indicates the time interval between the two occurrences. The smaller the value, the more synchronized the two are, and the closer the values of the first term are. Conversely, if the interval exceeds The value decays rapidly.
[0062] The preset time tolerance (200ms in this embodiment) defines the allowable time deviation range.
[0063] : An exponential function with the natural constant as its base.
[0064] By traversing all possible... and Combine and fill the entire matrix. The larger the value in the matrix, the higher the probability that the corresponding footstep and fall are the same physical event.
[0065] Step 402: Global optimal matching based on KM algorithm.
[0066] Simply counting the number of events or local matching can easily lead to misjudgments. In order to find the globally optimal interpretation, that is, to maximize the overall probability of association while ensuring that each event can be used at most once, this application introduces the bipartite graph maximum weight matching algorithm (Kuhn-Munkres Algorithm, or KM algorithm for short).
[0067] (1) Matrix completion: If (For example, more steps than falls), introduce virtual nodes to complete the matrix. The matrix is a square. The association weight between the virtual node and other nodes is set to a minimum value (0 in this embodiment), representing a no-match state.
[0068] (2) Path finding: The KM algorithm searches for a set of non-conflicting matching paths in the matrix (i.e., selecting elements from different rows and columns of the matrix) such that the sum of the weights of these elements is equal to the sum of the weights of the elements. It has reached its maximum value.
[0069] (3) Physical meaning: The maximum weight sum represents the total energy of events that can be explained by reasonable causal relationships within the current time window.
[0070] Step 403: Consistency score calculation and penalty mechanism.
[0071] To prevent the algorithm from forcibly matching poor data (such as forcibly matching weak vibrations to random signal fluctuations) and resulting in inflated scores, an effective matching threshold and a non-matching penalty are introduced when calculating the final score.
[0072] First, count the number of valid matches. That is, those matching weights ( This represents the matching threshold, which is 0.5 in this embodiment, and the number of pairs. Pairs that are selected by the KM algorithm but have extremely low weights are considered invalid matches.
[0073] Final walking signal consistency score The calculation formula is as follows: in, : Indicates the maximum total number of events within the current window. This represents the preset penalty coefficient (in this embodiment, the value is 0.2); The index set represents the valid matching pairs; s represents the index of the valid matching pairs. This represents the matching weight of the s-th valid match. For each unexplained isolated event (no match or poor match quality), the score is penalized by deducting points. Denominator normalization ensures... The range of values is Between. It should be noted that if If it equals 0, directly set It equals 0.
[0074] This step yields the walking signal consistency score. This score reflects not only the quality of the match (high weight) but also the completeness of the match (few isolated events). A score close to 1.0 indicates that the current behavior highly matches the characteristics of carrying a load while walking; a low score or even a negative score indicates chaotic behavior or a large number of fake signals.
[0075] The fifth step is to dynamically update the trust score based on the walking signal consistency score, and when the trust score meets the threshold condition, control the generation and destruction of the decryption key corresponding to the encrypted file data; and when the real-time RSSI value is continuously higher than the return threshold set based on the signal strength benchmark, trigger the destruction of the decryption key.
[0076] This step involves analyzing the walk signal consistency score. This translates instantaneous scores into continuous data access permissions. Given the complexity of the wireless environment, score fluctuations within a single time window are unavoidable. To improve terminal availability and prevent frequent lockouts due to occasional interference, a trust score pool mechanism with smoothing characteristics is established.
[0077] Step 501: Dynamic trust assessment based on points pool.
[0078] Allocate an integer variable in memory as the trust score. The initial value is set to 0. This trust score is used to quantify the terminal's cumulative trust level in the authenticity of the current offline behavior.
[0079] At the end of each sliding observation window, a consistency score is given based on the current walk signal. Update trust score: (1) Trust accumulation (bonus points): If the current rating Higher than the preset compliance threshold (In this embodiment, the value is 0.75, corresponding to the correlation between normal walking gait and signal), indicating that the behavioral characteristics in the current time period conform to the pattern of walking under load. At this time, an accumulation operation is performed to increase trust in the user: in, The preset maximum score is 100 in this embodiment. The step size for adding points (2 in this embodiment). This upper limit prevents the trust score from growing indefinitely, ensuring that continuous compliance behavior can quickly build trust, but also preventing subsequent anomalies from being ignored due to excessive trust; k represents the discrete time step; min() represents the minimization function.
[0080] (2) Trust decay (deduction): If the current rating Below the preset abnormal threshold (For example, 0.4) indicates an interruption or abnormality in behavioral characteristics (such as skateboard transport, remaining stationary). In this case, to prevent attackers from accumulating trust through intermittent deceptive behavior, a point deduction is performed: in, This represents the deduction step size (5 in this example). The deduction range is usually greater than the bonus range, reflecting the security principle that trust is difficult to establish and easy to break. The lower limit of integration is 0; max() represents the maximization function.
[0081] (3) Trust maintenance: If the score is between and Between these states, which are considered transitional states (such as brief pauses or signal fluctuations), the integral is maintained with a slight decay (the decay value is 1 in this embodiment) to maintain stability.
[0082] Step 502: Key generation and file decryption (permission granting).
[0083] To protect sensitive data, the terminal does not store decryption keys by default; archive files are stored in encrypted form. The system only generates temporary decryption capabilities when the trust score accumulates to a sufficiently high level.
[0084] Terminal real-time monitoring The value of: if and only if Exceeding the preset security confirmation threshold (In this embodiment, the value is 20, corresponding to approximately 5-10 seconds of continuous compliant walking.) When this occurs, the key generation process is triggered: (1) Call the Security Unit: The main processor sends a generation instruction to the built-in Security Unit (SE).
[0085] (2) Derivation of working key: The security element derives a volatile working key based on the preset root key. .
[0086] (3) Injecting volatile memory: Write to the protected random access memory (RAM) area of the system; writing to non-volatile storage media such as flash memory is strictly prohibited.
[0087] (4) Decryption and display: Application layer program call The encrypted archive data cached locally is decrypted using AES, and the decrypted plaintext is rendered onto the display screen for the escort personnel to view.
[0088] Step 503: Key erasure and privacy protection (access control circuit breaker) Once the trust chain breaks in an offline environment, the terminal must ensure that the data is immediately invisible to prevent leakage if the device falls into the hands of others.
[0089] In the key At any time after generation, if If the penalty points drop to 0, or if the device is detected returning to the vehicle (triggering a baseline update), the terminal immediately triggers the circuit breaker procedure: (1) Memory overwrite: The operating system calls the underlying driver to overwrite the memory. The RAM address area is overwritten multiple times with random numbers to ensure that the key cannot be recovered from residual magnetism.
[0090] (2) Handle release: Forcefully release all file handles and memory pointers associated with the key, interrupting any ongoing decryption operations.
[0091] (3) Screen blocking: Immediately turn off the display backlight or switch to a completely black lock screen to clear plaintext data in the video memory.
[0092] At this point, the archive data reverts to a completely unreadable ciphertext state due to the loss of the decryption key, until the next continuous compliance action is confirmed and sufficient trust points are accumulated again.
[0093] Based on the same inventive concept as the above methods, embodiments of this application also provide a multi-level encryption-based archive data access control system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described multi-level encryption-based archive data access control methods.
[0094] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.
[0095] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from its essential characteristics. Therefore, the embodiments described above should be considered exemplary and non-limiting in all respects; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some technical features, without causing the essence of the corresponding technical solutions to deviate from the scope of the technical solutions in the embodiments of this application, should all be included within the protection scope of this application.
Claims
1. A method for access control of archive data based on multi-level encryption, applied to portable terminals storing encrypted archive data, characterized in that: The method includes the following steps: Based on a preset number of consecutive RSSI values received by the terminal from a preset beacon device, a signal strength benchmark is determined. At the same time, based on the triaxial acceleration data collected by the terminal, an environmental noise vibration benchmark is obtained. One-dimensional synthetic modulus of triaxial acceleration data after removing gravity component is obtained. Foot impact judgment threshold is dynamically set based on the environmental noise vibration benchmark. The fluctuation characteristics of the modulus in the preset sliding observation window are compared with the foot impact threshold to generate foot vibration event. The continuous decrease change characteristics of RSSI value after detrending processing are analyzed to generate signal drop event. The differences between the corresponding attributes of foot vibration events and signal drop events are analyzed, an association weight matrix is constructed, and a bipartite graph maximum weight matching algorithm is used to search the association weight matrix to obtain matching paths and perform effective matching filtering. A non-matching penalty term is introduced, and a walking signal consistency score is calculated. The trust score is dynamically updated based on the walking signal consistency score, and the generation and destruction of the decryption key corresponding to the encrypted file data are controlled when the trust score meets the threshold condition; and the destruction of the decryption key is triggered when the real-time RSSI value is continuously higher than the return threshold set based on the signal strength benchmark.
2. The archive data access control method based on multi-level encryption as described in claim 1, characterized in that, The specific method for determining the signal strength benchmark is as follows: If the variance of the fluctuation of L consecutive RSSI values is lower than the preset stability threshold and the mean is higher than the effective connection threshold, then the mean of the L RSSI values is used as the signal strength benchmark, where L is the preset number.
3. The archive data access control method based on multi-level encryption as described in claim 1, characterized in that, The process of obtaining the environmental noise vibration reference is as follows: The root mean square value of the filtered acceleration data after a preset duration is used as the initial value of the environmental noise floor vibration reference, denoted as... The preset minimum effective vibration threshold is denoted as... ; like ,Will As the final environmental noise floor vibration benchmark; if The pre-calibrated default noise floor parameter is used as the final vibration amplitude benchmark.
4. The archive data access control method based on multi-level encryption as described in claim 1, characterized in that, The specific process for setting the foot impact detection threshold is as follows: The product of the preset signal-to-noise ratio coefficient and the environmental background noise vibration benchmark, plus the preset safety margin value, is used as the threshold for judging foot impact.
5. The archive data access control method based on multi-level encryption as described in claim 1, characterized in that, The specific process for generating the footstep vibration event is as follows: Based on all the magnitude values obtained in the sliding observation window, the peak corresponding to the maximum value that is greater than the foot impact judgment threshold is defined as a foot vibration event. The foot vibration event includes two attributes: the time of occurrence and the impact amplitude.
6. The archive data access control method based on multi-level encryption as described in claim 1, characterized in that, The signal drop event is specifically as follows: Based on all adjacent RSSI values after the trend has ended in the sliding observation window, the signal rate of change is calculated. When the signal rate of change is less than the preset signal drop threshold, and the time the signal maintains a downward trend exceeds the minimum drop duration, and the magnitude of the drop exceeds the preset minimum effective drop value, the corresponding sliding observation window is taken as the drop interval and defined as generating a signal drop event. The generated signal drop event includes two attributes: the time of the drop and the magnitude of the drop; the time of the drop is obtained by the midpoint of the drop interval; the magnitude of the drop is obtained by the difference between the signal value at the starting point and the signal value at the lowest point.
7. The archive data access control method based on multi-level encryption as described in claim 1, characterized in that, The construction of the correlation weight matrix is specifically as follows: The first in the correlation weight matrix Line number The specific elements of the column are: the first... The first footstep vibration event and the first Correlation weights among signal drop events The specific formula is as follows: in, This represents the time interval between the i-th foot vibration event and the j-th signal drop event; The default time tolerance is indicated by exp(); exp() represents an exponential function with the natural constant as the base.
8. The archive data access control method based on multi-level encryption as described in claim 1, characterized in that, The specific process for calculating the walk signal consistency score is as follows: The number of valid matches with a matching weight greater than a preset matching threshold in the matching path is denoted as . ; The maximum number of rows and columns in the correlation weight matrix is used as the maximum total number of events within the current window, denoted as . ; Walking signal consistency score The calculation formula is as follows: in, This indicates the preset penalty coefficient; The index set represents the valid matching pairs; s represents the index of the valid matching pairs. This represents the matching weight of the s-th valid matching pair.
9. The archive data access control method based on multi-level encryption as described in claim 1, characterized in that, The process of dynamically updating the trust score based on the consistency score of the walking signal is as follows: If the walking signal consistency score If the score exceeds the preset compliance threshold, a bonus will be awarded: ; If the walking signal consistency score If the score is below the preset abnormal threshold, a point deduction will be performed. ; Otherwise, the trust score will decrease by a preset decay rate; in, , Let these represent the trust integrals of the k-th and (k-1)-th sliding observation windows, respectively. The preset maximum score. The preset step size for adding points is defined by k, which represents the discrete time step; min() represents the function to be minimized. The preset deduction step size is greater than the addition step size; max() represents the maximization function.
10. A multi-level encryption-based archive data access control system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-9.