Vehicle track protection method and device based on geographical indistinguishability
By using hierarchical clustering and Kalman filtering techniques, a vehicle trajectory protection method is generated, which solves the problems of insufficient privacy protection and trajectory continuity in existing methods. It achieves location-based differentiated protection, thereby improving the strength of privacy protection and data availability.
Patent Information
- Application Number
- CN202511252994.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-03
- Publication Date
- 2025-12-09
AI Technical Summary
Existing vehicle trajectory protection methods have poor overall privacy protection capabilities, cannot be set according to the differences in privacy sensitivity at different locations, and disturbances will disrupt the continuity of the trajectory, affecting the accuracy of path recognition and traffic modeling.
A hierarchical clustering algorithm is used to generate a hierarchical clustering tree. The vehicle trajectory data clusters are merged using a merging function. The privacy protection level is generated based on the cluster density, and dynamic Kalman filtering is used to smooth the data using a Kalman model to dynamically adjust the privacy protection strength of the trajectory data.
It improves the efficiency of privacy budget utilization, enhances overall privacy protection capabilities, and enables different privacy sensitivity settings in different locations, maintaining the availability of trajectory data.
Smart Images
Figure CN121093384A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of privacy protection technology, and in particular relates to a method and device for protecting vehicle trajectories based on geographical indistinguishability. Background Technology
[0002] Vehicle trajectory data protection refers to the management or protection of spatiotemporal location information (such as time, latitude and longitude, speed, direction, etc.) generated by vehicles during operation, to prevent unauthorized access, misuse, leakage or tampering. It requires balancing data value and privacy security to ensure that the data is used effectively in a secure manner.
[0003] Existing solutions are mainly divided into two categories: trajectory disturbance solutions and trajectory replacement solutions.
[0004] 1. Trajectory perturbation scheme: Differential privacy random perturbation is usually applied to the data of each time step of the vehicle trajectory points.
[0005] First, a third party collects the data and processes it into a standardized format for easy management and storage. Then, for a specific trajectory, it is divided into several time-step trajectory point data points based on its timestamp. Each individual trajectory point data point is then randomly perturbed, i.e., randomly offset geographically by a certain distance. This offset distance is related to the allocated privacy budget. Regarding privacy budgets The smaller the value, the greater the disturbance and the better the protection; conversely, the larger the privacy budget... The larger the value, the smaller the perturbation and the weaker the protection. Privacy budgets are typically allocated across the entire trajectory, with each time step consuming a portion; the total cost is less than the total privacy budget. Differential privacy does not involve traditional encryption; instead, it uses data that is indistinguishable from the original data for publication, thus ensuring privacy protection without compromising data utility.
[0006] Differential privacy is originally defined as two adjacent datasets. and (Difference of at most one record), for any query function: ,algorithm satisfy - The conditions for differential privacy are: Specifically, for the trajectory data at each time step, Laplace noise is added to shift the data from the original trajectory data, thereby protecting the trajectory data. For trajectory data, differential privacy typically employs either trajectory-level differential privacy (DP) or point-level differential privacy (DP) strategies. Trajectory-level differential privacy: apply privacy mechanism to the whole trajectory as a whole, this scheme is more reasonable in practical application; Point-level differential privacy: add noise to each position point in the trajectory respectively, realize point-by-point protection, usually such scheme is not used because its cumulative privacy budget overhead is too large.
[0007] 2. Trajectory replacement scheme: use deep learning models such as LSTM, RNN, GAN, etc. to learn trajectory features, for real trajectory data, as input, through deep learning model processing, get a false trajectory data, false trajectory data and its original real trajectory data. Extract common path templates through historical trajectory clustering, location heat map analysis, etc. and add noise or fine-tune the path to generate pseudo-trajectory. For example, select common travel paths and construct multiple false trajectories without changing the macro distribution.
[0008] The existing scheme protects user privacy by adding random noise in essence, however, trajectory data has strong time sequence and spatial continuity, and disturbance will break the coherence of trajectory, speed constraint and turning mode, causing serious deformation of trajectory, affecting the accuracy of downstream tasks such as path recognition, trajectory clustering and traffic modeling; in addition, in trajectory-level or point-level differential privacy, privacy budget ε is the core parameter to measure the protection strength, but the continuous disturbance of multiple trajectory points will cause rapid consumption of privacy budget, resulting in poor overall privacy protection ability, and the existing method uses a fixed mechanism to disturb all trajectory points equally, ignoring the privacy sensitivity difference of different positions. For example: sensitive locations such as home and school should be protected, while highways or commercial areas can tolerate more public information. The overall privacy protection ability of the existing scheme is poor, and it cannot set the privacy sensitivity difference at different positions. SUMMARY
[0009] The application provides a vehicle trajectory protection method and device based on geographical indistinguishability, which can improve the protection strength of vehicle trajectory based on geographical indistinguishability.
[0010] To achieve the above object, the application provides a vehicle trajectory protection method based on geographical indistinguishability, comprising: Obtaining vehicle trajectory data, generating a hierarchical clustering tree according to the vehicle trajectory data using a preset hierarchical clustering algorithm, wherein the tree nodes of the hierarchical clustering tree represent vehicle trajectory data clustering clusters; Merging the vehicle trajectory data clustering clusters in the hierarchical clustering tree using a merging function to obtain final vehicle trajectory data clustering clusters, and generating a privacy protection level for each final vehicle trajectory data clustering cluster according to the clustering density of the final vehicle trajectory data clustering cluster; extract trajectory data of each time step in the vehicle trajectory data, allocate privacy budget to the trajectory data of each time step according to the final vehicle trajectory data cluster and the privacy protection level of each final vehicle trajectory data cluster, and obtain scrambled trajectory data of each time step; Obtain trajectory speed, acceleration and coordinate information of the scrambled trajectory data of each time step, input the trajectory speed, acceleration and coordinate information into a Kalman model to perform dynamic Kalman filtering smoothing on the scrambled trajectory data of each time step, and complete privacy protection of the vehicle trajectory data.
[0011] To solve the above problems, the application also provides a vehicle trajectory protection device based on geographical indistinguishability, which comprises: A data clustering module is configured to obtain vehicle trajectory data, and generate a hierarchical clustering tree by using a preset hierarchical clustering algorithm according to the vehicle trajectory data, wherein a tree node of the hierarchical clustering tree represents a vehicle trajectory data cluster; A privacy budget allocation module is configured to merge vehicle trajectory data clusters in the hierarchical clustering tree by using a merging function to obtain final vehicle trajectory data clusters, generate a privacy protection level of each final vehicle trajectory data cluster according to the clustering density of the final vehicle trajectory data cluster, extract trajectory data of each time step in the vehicle trajectory data, and allocate privacy budget to the trajectory data of each time step according to the final vehicle trajectory data cluster and the privacy protection level of each final vehicle trajectory data cluster to obtain scrambled trajectory data of each time step; A Kalman filtering smoothing module is configured to obtain trajectory speed, acceleration and coordinate information of the scrambled trajectory data of each time step, input the trajectory speed, acceleration and coordinate information into a Kalman model to perform dynamic Kalman filtering smoothing on the scrambled trajectory data of each time step, and complete privacy protection of the vehicle trajectory data.
[0012] To solve the above problems, the application also provides an electronic device, which comprises: at least one processor; and a memory in communication connection with the at least one processor; wherein The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the vehicle trajectory protection method based on geographical indistinguishability described above.
[0013] To solve the above problems, the application further provides a computer readable storage medium, wherein at least one computer program is stored in the computer readable storage medium, and the at least one computer program is executed by a processor in an electronic device to implement the vehicle trajectory protection method based on geographical indistinguishability.
[0014] The application obtains vehicle trajectory data, generates a hierarchical clustering tree by using a hierarchical clustering algorithm, and merges vehicle trajectory data clusters in the hierarchical clustering tree by using a merging function. The hierarchical clustering algorithm and the merging function can automatically divide a space region from fine granularity to coarse granularity, can indirectly infer the privacy protection level of a region according to the aggregation degree of a trajectory point, and additionally, the trajectory data of each time step is allocated a privacy budget according to the final vehicle trajectory data cluster and the privacy protection level of each final vehicle trajectory data cluster, and is scrambled according to the privacy protection level of each final vehicle trajectory data cluster, thereby improving the utilization efficiency of the privacy budget and enhancing the overall privacy protection capability. The final vehicle trajectory data cluster can be introduced to set the privacy sensitivity difference at different positions, and in addition, the Kalman filter is used to dynamically smooth the scrambled trajectory, thereby improving the usability of the disturbed trajectory data. BRIEF DESCRIPTION OF DRAWINGS
[0015] Figure 1 A flowchart of a vehicle trajectory protection method based on geographical indistinguishability is provided for an embodiment of the application. Figure 2 A structure flowchart of an example of a vehicle trajectory protection method based on geographical indistinguishability is provided for an embodiment of the application. Figure 3 A privacy leakage risk diagram of different privacy budgets under a truncation radius of 10 meters and 100 meters of a vehicle trajectory protection method based on geographical indistinguishability is provided for an embodiment of the application. Figure 4 A privacy leakage risk diagram of different privacy budgets under a truncation radius of 1000 meters of a vehicle trajectory protection method based on geographical indistinguishability is provided for an embodiment of the application. Figure 5 A functional module diagram of a vehicle trajectory protection device based on geographical indistinguishability is provided for an embodiment of the application. Figure 6 A structure diagram of an electronic device for implementing the vehicle trajectory protection method based on geographical indistinguishability is provided for an embodiment of the application.
[0016] The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION
[0017] It is to be understood that the specific embodiments described herein are merely illustrative of the present application and do not limit the scope of the application.
[0018] An embodiment of the present application provides a vehicle trajectory protection method based on geographical indistinguishability. The execution subject of the vehicle trajectory protection method based on geographical indistinguishability includes but is not limited to at least one of electronic devices capable of being configured to execute the method provided by the embodiment of the present application, such as a server and a terminal. In other words, the vehicle trajectory protection method based on geographical indistinguishability can be executed by software or hardware installed in a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to a single server, a server cluster, a cloud server, or a cloud server cluster, etc. The server can be a stand-alone server, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content distribution networks (CDN), and big data and artificial intelligence platforms, etc. basic cloud computing services.
[0019] Reference Figure 1 FIG. 1 shows a flowchart of a vehicle trajectory protection method based on geographical indistinguishability provided by an embodiment of the present application. In this embodiment, the vehicle trajectory protection method based on geographical indistinguishability includes: S1, obtaining vehicle trajectory data, and generating a hierarchical clustering tree according to the vehicle trajectory data using a preset hierarchical clustering algorithm, wherein a tree node of the hierarchical clustering tree represents a vehicle trajectory data clustering cluster.
[0020] It can be understood that vehicle trajectory data refers to a data set recording the continuous motion state of a vehicle within a period of time, and the vehicle trajectory data contains a series of time-sequentially arranged data points, each of which records the position of the vehicle at a specific time step and related motion state information.
[0021] It can be understood that the preset hierarchical clustering algorithm refers to an algorithm structure for constructing a tree structure for hierarchical clustering, and the preset hierarchical clustering algorithm can be an HDBSCAN (Hierarchical Density-Based Spatial Clustering of Applications with Noise) algorithm.
[0022] Specifically, the step of generating a hierarchical clustering tree according to the vehicle trajectory data using a preset hierarchical clustering algorithm includes: using the preset hierarchical clustering algorithm to identify the hierarchical structure in the vehicle trajectory data, and calculating the mutual reachable distance of the trajectory data at each time step in the vehicle trajectory data; According to the mutual reachable distance, a mutual reachable distance matrix is constructed, and a minimum spanning tree is constructed according to the mutual reachable distance matrix, and the minimum spanning tree is used to generate a hierarchical clustering tree by using a method of deleting edges in reverse.
[0023] It can be understood that the mutual reachable distance refers to the mutual reachable distance between the trajectory data of two time steps.
[0024] It can be understood that the minimum spanning tree refers to a tree connecting all vertices in a complete graph constructed by taking the trajectory data of each time step as a vertex and the weight of the mutual reachable distance as an edge, and the sum of the weights of the edges is the minimum.
[0025] Further, after the minimum spanning tree is used to generate a hierarchical clustering tree by using a method of deleting edges in reverse, the hierarchical clustering tree is further compressed to obtain a standard hierarchical clustering tree.
[0026] Further, the compression of the hierarchical clustering tree to obtain a standard hierarchical clustering tree comprises: starting from the root node of the hierarchical clustering tree, traversing the hierarchical clustering tree, and deleting tree nodes in the hierarchical clustering tree according to a preset minimum cluster size; wherein the deletion of the tree nodes in the hierarchical clustering tree according to the preset minimum cluster size comprises: when the number of samples of both child nodes under the tree node is greater than or equal to the preset minimum cluster size, the tree node and the two child nodes under the tree node are retained; when the number of samples of one of the two child nodes under the tree node is less than the preset minimum cluster size, the child node is deleted, and the other child node is retained; when the number of samples of both child nodes under the tree node is less than the preset minimum cluster size, the two child nodes under the tree node are deleted.
[0027] Exemplarily, vehicle trajectory data is obtained, and a hierarchical clustering tree is generated from the vehicle trajectory data by using a preset hierarchical clustering algorithm. The following implementation steps can be used: Step 1, privacy budget allocation: wherein HDBSCAN is a density adaptive hierarchical clustering algorithm, which discriminates clusters by setting a mutual reachable distance: wherein the core distance of the sample point is the minimum radius that makes its neighborhood contain at least samples; The preset minimum cluster size is obtained; a minimum spanning tree (MST) is constructed based on a mutual reachability metric distance, a hierarchical clustering tree is generated by reversely deleting edges, each node in the tree represents a potential cluster, and an edge weight corresponds to a separation distance between clusters.
[0028] S2, merging the vehicle trajectory data clustering clusters in the hierarchical clustering tree by using a merging function to obtain final vehicle trajectory data clustering clusters, and generating a privacy protection level of each final vehicle trajectory data clustering cluster according to a clustering density of the final vehicle trajectory data clustering cluster.
[0029] It can be understood that the merging function is a function for post-processing clustering results, and generally optimizes and merges preliminary clustering results generated by a clustering algorithm (HDBSCAN), and the merging function can be a merge function.
[0030] Exemplarily, the merging function is used to merge the vehicle trajectory data clustering clusters in the hierarchical clustering tree, and the following implementation steps can be adopted: The merging function is used to judge the level and overlapping degree of two privacy areas, and the merging operation is performed, and the specific scheme is as follows: wherein, is the time step of the privacy area , and is the privacy level of the privacy area at the time step , and is the overlapping degree of the privacy area and the privacy area , and is the overlapping threshold.
[0031] For the areas A and B, if the two privacy areas overlap more than the threshold, the two areas are merged, and the privacy strength is set to be higher.
[0032] As an embodiment of the present application, the preliminary clustering results generated by the clustering algorithm (HDBSCAN) can be optimized and merged by using the following formula: wherein, is the vehicle trajectory data, is the privacy area.
[0033] For the privacy area, the following formula can be used to merge the privacy area to improve the efficiency of subsequent queries: Finally, the real-time privacy level area is obtained Its corresponding privacy protection level ,in, For the first A private area, For the first A private area.
[0034] Understandably, vehicle trajectory data consists of a continuous sequence of states, representing the geographical coordinates of the vehicle at different timestamps, denoted as... S3. Extract the trajectory data for each time step from the vehicle trajectory data, allocate the privacy budget for the trajectory data of each time step according to the final vehicle trajectory data cluster and the privacy protection level of each final vehicle trajectory data cluster, and obtain the scrambled trajectory data for each time step.
[0035] Specifically, before allocating the privacy budget for the trajectory data at each time step based on the final vehicle trajectory data clusters and the privacy protection level of each final vehicle trajectory data cluster, the following steps are also included: Set a cutoff radius and apply a planar Laplace perturbation to the trajectory data at a single time step. For the trajectory data at a certain time step... The perturbation formula is as follows: in, Represented as arrive Euclidean distance, For privacy budget, The first normalization constant, For trajectory data The disturbance.
[0036] Understandably, since the disturbance occurs in polar coordinates, the disturbance distribution is radially symmetrical and depends only on the cutoff radius.
[0037] For example, the trajectory data at each time step is scrambled for each time step by allocating a privacy budget based on the final vehicle trajectory data cluster and the privacy protection level of each final vehicle trajectory data cluster, using the following implementation steps: by Acceptable privacy area with maximum cutoff radius For trajectory data that does not meet the requirements of time step The specific mechanism for rejection is as follows: in, The second normalization constant, For mechanism trajectory data the probability that the perturbation is .
[0038] where the second normalization constant is: Considering the geographic indistinguishability, for a privacy leakage upper bound , for any two locations , , in the privacy region the perturbation mechanism satisfies: Specifically, a circular region with a truncated radius R is constructed as the perturbation allowed range, if the trajectory point after random perturbation falls outside the range, the perturbation is resampled, that is, perturbed again until it falls within the range. Since the limit is performed, considering that the maximum leakage occurs at the center point, i.e., the publishing point , at this time: At this time, for the rejected sampled point, there is a privacy leakage upper bound , which satisfies: It can be understood that for the rejected sampled point, there is a privacy leakage upper bound , that is, for the truncated radius , when the privacy budget is set to , there is a leakage risk of .
[0039] S4, obtain the trajectory speed, acceleration and coordinate information of each time step of the perturbed trajectory data, input the trajectory speed, acceleration and coordinate information into the Kalman model to perform dynamic Kalman filtering smoothing on each time step of the perturbed trajectory data, and complete the privacy protection of the vehicle trajectory data.
[0040] It can be understood that the trajectory speed refers to the moving speed of the perturbed trajectory point.
[0041] It can be understood that the coordinate information refers to the false coordinate information of the perturbed trajectory point.
[0042] It can be understood that the acceleration refers to the speed change rate of the perturbed trajectory point.
[0043] It can be understood that the Kalman model refers to an optimal recursive data estimation algorithm that uses a series of observation data containing noise to estimate dynamic states and minimize the mean square error of the estimation.
[0044] Exemplarily, the trajectory speed, acceleration and coordinate information are input into the Kalman model to dynamically Kalman filter and smooth the perturbed trajectory data of each time step, complete the privacy protection of the vehicle trajectory data, and the following implementation steps can be used: Step 1: input trajectory speed, acceleration and coordinate information The trajectory speed, acceleration and coordinate information are obtained by code calculation or reading original vehicle trajectory data and input into the Kalman model, and the specific steps are as follows: Define the state transition model: wherein, is a state vector, is process noise generated by motion change, is a trajectory speed, F is a state transition matrix, and M is a unit speed model, which is set as follows: Step 2: set the observation covariance matrix According to the obtained privacy area and the privacy protection level , the observation covariance matrix R is determined, which acts on the Kalman gain : wherein is a predicted state covariance matrix, is an observation matrix, is a transpose of the observation matrix.
[0045] It can be understood that in the embodiment of the application, the covariance matrix of the high sensitive area is set to be larger, the observation noise of these areas is larger, which means that the trajectory points of these areas need to be smoothed more strongly, the observation value is more untrusted, and the estimated state more depends on the model prediction; on the contrary, for the low sensitive area, the covariance is set to be smaller.
[0046] The application obtains vehicle trajectory data, generates a hierarchical clustering tree by using a hierarchical clustering algorithm, and merges vehicle trajectory data clusters in the hierarchical clustering tree by using a merging function. The hierarchical clustering algorithm and the merging function can automatically divide the space region from fine granularity to coarse granularity, can indirectly infer the privacy protection level of the region according to the aggregation degree of the trajectory points, and additionally, the trajectory data of each time step is allocated privacy budget according to the final vehicle trajectory data cluster and the privacy protection level of each final vehicle trajectory data cluster, and is scrambled according to the privacy protection level of each final vehicle trajectory data cluster, thereby improving the utilization efficiency of the privacy budget and improving the overall privacy protection capability. The introduction of the final vehicle trajectory data cluster can realize the setting of the privacy sensitivity difference in different positions, and in addition, the Kalman filter is used to dynamically smooth the scrambled trajectory, thereby improving the availability of the disturbed trajectory data.
[0047] Referring to Figure 2 As shown in the structure flow chart of one example of the vehicle trajectory protection method based on geographical indistinguishability provided by an embodiment of the application.
[0048] Referring to Figure 3 As shown in the privacy leakage risk diagram of different privacy budgets under a truncation radius of 10 meters and 100 meters of the vehicle trajectory protection method based on geographical indistinguishability provided by an embodiment of the application.
[0049] Referring to Figure 4 As shown in the privacy leakage risk diagram of different privacy budgets under a truncation radius of 1000 meters of the vehicle trajectory protection method based on geographical indistinguishability provided by an embodiment of the application.
[0050] As Figure 5 shown, it is a functional module diagram of a vehicle trajectory protection device based on geographical indistinguishability provided by an embodiment of the application.
[0051] The vehicle trajectory protection device 100 based on geographical indistinguishability provided by the application can be installed in an electronic device. According to the functions implemented, the vehicle trajectory protection device 100 based on geographical indistinguishability can include a data clustering module 101, a privacy budget allocation module 102, and a Kalman filter smoothing module 103.
[0052] The module described in the application can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete a fixed function, and is stored in the memory of the electronic device.
[0053] In this embodiment, the functions of each module / unit are as follows: The data clustering module 101 is configured to acquire vehicle trajectory data, and generate a hierarchical clustering tree according to the vehicle trajectory data by using a preset hierarchical clustering algorithm, wherein a tree node of the hierarchical clustering tree represents a vehicle trajectory data clustering cluster.
[0054] The privacy budget allocation module 102 is configured to merge the vehicle trajectory data clustering clusters in the hierarchical clustering tree by using a merging function to obtain final vehicle trajectory data clustering clusters, generate a privacy protection level of each final vehicle trajectory data clustering cluster according to a clustering density of the final vehicle trajectory data clustering cluster, extract trajectory data of each time step in the vehicle trajectory data, and allocate a privacy budget to the trajectory data of each time step according to the final vehicle trajectory data clustering cluster and the privacy protection level of each final vehicle trajectory data clustering cluster to obtain scrambled trajectory data of each time step.
[0055] The Kalman filter smoothing module 103 is configured to acquire trajectory speed, acceleration and coordinate information of the scrambled trajectory data of each time step, input the trajectory speed, acceleration and coordinate information into a Kalman model to perform dynamic Kalman filter smoothing on the scrambled trajectory data of each time step, and complete privacy protection of the vehicle trajectory data.
[0056] As shown in Figure 6 FIG. 1 is a structural schematic diagram of an electronic device for implementing a vehicle trajectory protection method based on geographical indistinguishability according to an embodiment of the present application.
[0057] The electronic device can include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and can further include a computer program stored in the memory 11 and executable on the processor 10, such as a vehicle trajectory protection method program based on geographical indistinguishability.
[0058] In some embodiments, the processor 10 can be composed of integrated circuits, for example, can be composed of a single packaged integrated circuit, or can be composed of multiple packaged integrated circuits with the same function or different functions, including one or more combinations of central processing units (CPU), microprocessors, digital processing chips, graphics processors and various control chips. The processor 10 is the control core of the electronic device, which connects all components of the electronic device by using various interfaces and lines, executes or runs programs or modules stored in the memory 11 (for example, executes a vehicle trajectory protection method program based on geographical indistinguishability), and calls data stored in the memory 11 to perform various functions of the electronic device and process data.
[0059] The memory 11 includes at least one type of readable storage medium, such as a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. The memory 11 can be an internal storage unit of an electronic device in some embodiments, such as a mobile hard disk of the electronic device. The memory 11 can also be an external storage device of the electronic device in other embodiments, such as a plug-in mobile hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device. Further, the memory 11 can include both an internal storage unit and an external storage device of the electronic device. The memory 11 can be used not only to store application software and various data installed in the electronic device, such as a code of a vehicle trajectory protection method based on geographical indistinguishability, but also to temporarily store data that has been output or will be output.
[0060] The communication bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable connection and communication between the memory 11 and at least one processor 10, etc.
[0061] The communication interface 13 is used for communication between the electronic device and other devices, including a network interface and a user interface. Optionally, the network interface can include a wired interface and / or a wireless interface (e.g., a WI-FI interface, a Bluetooth interface, etc.), which is usually used to establish a communication connection between the electronic device and other electronic devices. The user interface can be a display, an input unit (such as a keyboard), and optionally, the user interface can also be a standard wired interface, a wireless interface. Optionally, in some embodiments, the display can be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, etc. The display can also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device and to display a visualized user interface.
[0062] Figure 6 Only an electronic device with components is shown, and those skilled in the art can understand that, Figure 6The illustrated structure does not constitute a limitation on the electronic device, which can include fewer or more components than those shown, or combine certain components, or arrange the components differently.
[0063] For example, although not shown, the electronic device can also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, so that functions such as charge management, discharge management, and power consumption management can be realized through the power management device. The power supply can also include one or more direct current or alternating current power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and any other components. The electronic device can also include various sensors, Bluetooth modules, Wi-Fi modules, and the like, which are not described here.
[0064] It should be understood that the embodiments are only for illustration and are not limited in the scope of the patent application by this structure.
[0065] The program stored in the memory 11 of the electronic device is a combination of a plurality of instructions, which, when executed in the processor 10, can implement: Obtaining vehicle trajectory data, and generating a hierarchical clustering tree according to the vehicle trajectory data using a preset hierarchical clustering algorithm, wherein a tree node of the hierarchical clustering tree represents a vehicle trajectory data cluster; Merging the vehicle trajectory data clusters in the hierarchical clustering tree using a merging function to obtain final vehicle trajectory data clusters, and generating a privacy protection level for each final vehicle trajectory data cluster according to the clustering density of the final vehicle trajectory data cluster; Extracting trajectory data at each time step from the vehicle trajectory data, and performing privacy budget allocation on the trajectory data at each time step according to the final vehicle trajectory data clusters and the privacy protection level of each final vehicle trajectory data cluster to obtain scrambled trajectory data at each time step; Obtaining trajectory speed, acceleration, and coordinate information of the scrambled trajectory data at each time step, and inputting the trajectory speed, acceleration, and coordinate information into a Kalman model to perform dynamic Kalman filter smoothing on the scrambled trajectory data at each time step to complete privacy protection of the vehicle trajectory data.
[0066] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to the description of the related steps in the corresponding embodiments of the accompanying drawings, which is not described here.
[0067] Further, the modules / units integrated in the electronic device 1 are stored in a computer readable storage medium if they are implemented in the form of software function units and sold or used as independent products. The computer readable storage medium can be volatile or non-volatile. For example, the computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM).
[0068] The application further provides a computer readable storage medium, which stores a computer program, and the computer program can realize the following when executed by a processor of an electronic device: Obtaining vehicle trajectory data, and generating a hierarchical clustering tree according to the vehicle trajectory data by using a preset hierarchical clustering algorithm, wherein a tree node of the hierarchical clustering tree represents a vehicle trajectory data cluster; Merging the vehicle trajectory data clusters in the hierarchical clustering tree by using a merging function to obtain final vehicle trajectory data clusters, and generating a privacy protection level of each final vehicle trajectory data cluster according to a clustering density of the final vehicle trajectory data cluster; Extracting trajectory data of each time step in the vehicle trajectory data, and performing privacy budget allocation on the trajectory data of each time step according to the final vehicle trajectory data clusters and the privacy protection level of each final vehicle trajectory data cluster to obtain scrambled trajectory data of each time step; Obtaining trajectory speed, acceleration and coordinate information of the scrambled trajectory data of each time step, inputting the trajectory speed, acceleration and coordinate information into a Kalman model to perform dynamic Kalman filtering smoothing on the scrambled trajectory data of each time step, and completing privacy protection of the vehicle trajectory data.
[0069] In the several embodiments provided in the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative, and the division of the modules is only a logical function division. In actual implementation, another division manner can be used.
[0070] The modules described as separate components can or can not be physically separated, and the components displayed as modules can or can not be physical units, i.e., can be located in one place or distributed on multiple network units. According to actual needs, part or all of the modules can be selected to achieve the purpose of the present embodiment.
[0071] In addition, each functional module in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware, or in the form of hardware plus software function module.
[0072] It is obvious for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and the present application can be realized in other specific forms without departing from the spirit or essential characteristics of the present application.
[0073] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, the scope of the present application is defined by the appended claims rather than the above description, and therefore all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present application. Any additional reference signs in the claims should not be considered as limiting the claims to which they relate.
[0074] The blockchain referred to in the present application is a new application mode of distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm and other computer technologies. Blockchain, in essence, is a decentralized database, which is a series of data blocks associated using cryptography. Each data block contains information about a batch of network transactions, which is used to verify the validity of the information (anti-fake) and generate the next block. The blockchain can include a blockchain underlying platform, a platform product service layer, and an application service layer.
[0075] The embodiments of the present application can acquire and process related data based on artificial intelligence technology. Among them, artificial intelligence (Artificial Intelligence, AI) is to use digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. Theory, method, technology and application system.
[0076] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The plurality of units or devices stated in the system claims can also be realized by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, not any particular order.
[0077] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application.
Claims
1. A vehicle trajectory protection method based on geographical indistinguishability, characterized in that, The method includes: Acquire vehicle trajectory data, and generate a hierarchical clustering tree based on the vehicle trajectory data using a preset hierarchical clustering algorithm. The tree nodes of the hierarchical clustering tree represent the clusters of vehicle trajectory data. The vehicle trajectory data clusters in the hierarchical clustering tree are merged using a merging function to obtain the final vehicle trajectory data clusters. The privacy protection level of each final vehicle trajectory data cluster is generated based on the clustering density of the final vehicle trajectory data clusters. The trajectory data for each time step in the vehicle trajectory data is extracted. The trajectory data for each time step is then allocated a privacy budget based on the final vehicle trajectory data cluster and the privacy protection level of each final vehicle trajectory data cluster, resulting in scrambled trajectory data for each time step. The system acquires the trajectory velocity, acceleration, and coordinate information of the scrambled trajectory data at each time step. The trajectory velocity, acceleration, and coordinate information are then input into the Kalman model to perform dynamic Kalman filtering smoothing on the scrambled trajectory data at each time step, thus achieving privacy protection for the vehicle trajectory data.
2. The vehicle trajectory protection method based on geographical indistinguishability as described in claim 1, characterized in that, The step of generating a hierarchical clustering tree based on vehicle trajectory data using a preset hierarchical clustering algorithm includes: The hierarchical structure in vehicle trajectory data is identified using a pre-defined hierarchical clustering algorithm, and the reachability distance between trajectory data at each time step in the vehicle trajectory data is calculated. Construct a mutual reachability matrix based on the mutual reachability distance, and then construct a minimum spanning tree based on the mutual reachability distance matrix. Finally, use the minimum spanning tree to generate a hierarchical clustering tree by deleting edges in reverse.
3. The vehicle trajectory protection method based on geographical indistinguishability as described in claim 2, characterized in that, After generating a hierarchical clustering tree from the minimum spanning tree using the method of reverse edge deletion, the method further includes: compressing the hierarchical clustering tree to obtain a standard hierarchical clustering tree.
4. The vehicle trajectory protection method based on geographical indistinguishability as described in claim 3, characterized in that, The process of compressing the hierarchical clustering tree to obtain a standard hierarchical clustering tree includes: The hierarchical clustering tree is traversed starting from the root node, and tree nodes in the hierarchical clustering tree are deleted according to the preset minimum cluster size. The process of deleting tree nodes in the hierarchical clustering tree based on a preset minimum cluster size includes: If the number of samples in both child nodes under a tree node is greater than or equal to the preset minimum cluster size, retain the tree node and its two child nodes. If the number of samples in one of the two child nodes under a tree node is less than the preset minimum cluster size, delete the child node and keep the other child node. If the number of samples in both child nodes of a tree node is less than the preset minimum cluster size, delete both child nodes of that tree node.
5. The vehicle trajectory protection method based on geographical indistinguishability as described in any one of claims 1-4, characterized in that, Before allocating the privacy budget based on the final vehicle trajectory data clusters and the privacy protection level of each final vehicle trajectory data cluster, the following steps are also included: Set a cutoff radius and apply a planar Laplace perturbation to the trajectory data at a single time step. For the trajectory data at a certain time step... The perturbation formula is as follows: in, Represented as arrive Euclidean distance, For privacy budget, The first normalization constant, For trajectory data The disturbance.
6. The vehicle trajectory protection method based on geographical indistinguishability as described in any one of claims 1-4, characterized in that, The step of setting a cutoff radius and perturbing the trajectory data of a single time step with a planar Laplace's perturbation includes: setting an acceptable privacy region based on the maximum value of the cutoff radius, and using a preset perturbation mechanism to perturb the trajectory data outside the acceptable privacy region to the acceptable privacy region, thereby obtaining scrambled trajectory data for a single time step.
7. A vehicle trajectory protection device based on geographical indistinguishability, characterized in that, The apparatus is used to implement the vehicle trajectory protection method based on geographic indistinguishability as described in any one of claims 1 to 6, and the apparatus comprises: The data clustering module is used to acquire vehicle trajectory data and generate a hierarchical clustering tree based on the vehicle trajectory data using a preset hierarchical clustering algorithm. The tree nodes of the hierarchical clustering tree represent the vehicle trajectory data clusters. The privacy budget allocation module is used to merge vehicle trajectory data clusters in the hierarchical clustering tree using a merging function to obtain the final vehicle trajectory data clusters. Based on the cluster density of the final vehicle trajectory data clusters, a privacy protection level is generated for each final vehicle trajectory data cluster. The module extracts trajectory data for each time step from the vehicle trajectory data and allocates a privacy budget for the trajectory data at each time step based on the final vehicle trajectory data clusters and the privacy protection level of each final vehicle trajectory data cluster, thus obtaining scrambled trajectory data for each time step. The Kalman filter smoothing module is used to acquire the trajectory velocity, acceleration, and coordinate information of the scrambled trajectory data at each time step. The trajectory velocity, acceleration, and coordinate information are input into the Kalman model to perform dynamic Kalman filter smoothing on the scrambled trajectory data at each time step, thus completing the privacy protection of vehicle trajectory data.
8. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the geographically indistinguishable vehicle trajectory protection method as described in any one of claims 1 to 6.
9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the vehicle trajectory protection method based on geographical indistinguishability as described in any one of claims 1 to 6.