Differential privacy location protection method based on dynamic attenuation noise field

Abstract

The invention relates to a differential privacy position protection method based on a dynamic attenuation noise field, and the method comprises the steps: obtaining a plurality of candidate regions based on a user track, recognizing different sensitive regions from the plurality of candidate regions based on the behavior characteristics of a user, and calculating the dynamic sensitivity of each sensitive region; wherein the behavior characteristics of the user comprise a stay identifier and access time of the user in the candidate area; constructing a three-dimensional dynamic noise model for generating a dynamically attenuated space-time noise field; the method comprises the following steps: acquiring a position coordinate of a target position in a sensitive area and a corresponding timestamp, generating a corresponding noise amplitude by adopting a three-dimensional dynamic noise model, generating a two-dimensional position offset based on the noise amplitude by adopting a Laplace mechanism, and disturbing the position coordinate of the target position based on the two-dimensional position offset. According to the scheme, adaptive adjustment of the noise amplitude can be realized, and collaborative optimization of strong protection of a high-sensitivity area and high precision of a low-sensitivity area is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of information privacy protection, and particularly relates to a differential privacy location protection method based on a dynamic attenuation noise field. BACKGROUND

[0002] In location-based service applications, such as online car hailing trajectory, take-out delivery, and the like, a traditional differential privacy technology usually adopts a fixed-strength noise addition mechanism. This scheme adds the same amplitude of disturbance to all geographic locations without distinction, which results in insufficient noise intensity in sensitive areas and cannot effectively prevent the leakage of personal sensitive locations such as home addresses and frequently visited medical points; and over-adds noise to high-activity areas such as commercial districts and transportation hubs, causing path planning confusion and scheduling efficiency decline. This "one-size-fits-all" noise strategy is difficult to balance privacy protection strength and data utility.

[0003] Secondly, the traditional scheme ignores the influence of data timeliness, and the excessive protection of historical data reduces the long-term analysis value of the data. The static noise model cannot adapt to the dynamic changes of the city. The emergence of emerging residential areas and other sensitive places causes the original noise configuration to fail, and the manual adjustment mechanism responds slowly, causing privacy protection vulnerabilities. Finally, in large cities, fixed noise parameters are difficult to adapt to regional population density differences, and there is a problem of insufficient noise intensity in high-density areas.

[0004] Most current location-based services rely on large-scale user location data to support business decisions. However, raw location data contains sensitive information about users, and direct publication can lead to user privacy leakage. In order to protect user privacy, services will add noise to achieve individual privacy protection when publishing data through differential privacy technology. However, in the high-dimensional and dynamic location data scenario, traditional differential privacy is insufficient in protecting sensitive areas, and the same noise intensity is used for high-sensitivity areas and low-sensitivity areas, which cannot balance privacy and utility. Secondly, there is a conflict of data timeliness, and the same privacy budget is used to protect new and old data, which reduces the long-term analysis value of historical data, and insufficient protection of new data causes real-time privacy leakage. Finally, there is a lack of dynamic adjustment, and current location services mostly use manual configuration of static noise parameters, which cannot adapt to the emergence of emerging sensitive places.

[0005] The traditional technical solution adopts uniform noise injection, adds noise of fixed strength to the position coordinates through Laplace mechanism or Gaussian mechanism, cannot balance the privacy protection strength and utility, and is prone to cause insufficient protection strength in high-sensitive areas and excessive distortion of data in low-sensitive areas. Some solutions adopt local sensitive hash index technology, map the location data to Hamming space and construct a single hash table index. The defect is that the hash table structure causes insufficient hash collision probability, significantly reduces the recall rate of the neighbor query. The index update mechanism is rigid, and when the user behavior pattern changes (such as adding a new frequently visited place), the index needs to be fully reconstructed, and the calculation overhead increases in square with the data size, which is difficult to support dynamic scenarios. Some solutions use Bayesian network modeling to model the mutual information dependency between attributes, and allocate differentiated privacy budgets according to the correlation strength. However, the sensitive area recognition depends on the pre-set semantic label and cannot automatically discover emerging sensitive areas. At the same time, the budget allocation strategy ignores the time dimension decay requirement, and applies the same protection strength to historical data and new data, causing unnecessary loss of historical data analysis value.

[0006] Therefore, the disadvantages of the prior art mainly include the following three points: 1. Low spatial sensitivity differentiation: insufficient protection in high-sensitive areas and excessive distortion in low-sensitive areas.

[0007] 2. Dependence on artificial pre-set geographic fence or semantic label: lagging recognition of emerging sensitive areas and high maintenance cost.

[0008] 3. Ignoring the time effectiveness: applying high-strength noise to historical data and new data without distinction, resulting in unnecessary loss of historical data value.

[0009] Therefore, in the field of location data privacy protection, the traditional differential privacy method causes insufficient protection in high-sensitive areas such as residential areas and excessive distortion in low-sensitive areas such as commercial areas due to the use of fixed noise strength. At the same time, the existing solutions rely on artificial pre-set sensitive area boundaries, which are difficult to adapt to urban dynamic development. SUMMARY

[0010] The technical problem to be solved by the present application is to provide a differential privacy location protection method based on a dynamic decay noise field.

[0011] To achieve the above-mentioned application purpose, the present application provides a differential privacy location protection method based on a dynamic decay noise field, comprising the following steps: S1. Obtain a plurality of candidate areas based on user trajectories, and identify different sensitive areas from the plurality of candidate areas based on the behavior characteristics of the user, and calculate the dynamic sensitivity of each sensitive area; wherein the behavior characteristics of the user include: the stay identifier and access time of the user in the candidate area; S2. constructing a three-dimensional dynamic noise model for generating a dynamic attenuated spatio-temporal noise field; wherein the three-dimensional dynamic noise model is used to generate different noise amplitudes for different position coordinates and different time stamps in the sensitive region; S3. obtaining the position coordinates of a target position in the sensitive region and the corresponding time stamp, and generating the corresponding noise amplitude by using the three-dimensional dynamic noise model, and generating a two-dimensional position offset by using a Laplace mechanism based on the noise amplitude, and perturbing the position coordinates of the target position based on the two-dimensional position offset, and outputting the perturbed position coordinates.

[0012] According to an aspect of the present application, in step S1, in the step of obtaining a plurality of candidate regions based on user trajectories, the step comprises: collecting a series of trajectory points in the user trajectories, and obtaining a spatio-temporal feature vector of each trajectory point; clustering the trajectory points based on the spatio-temporal feature vectors and using a clustering method to generate a plurality of candidate regions.

[0013] According to an aspect of the present application, in step S1, in the step of identifying different sensitive regions from a plurality of candidate regions based on the behavior characteristics of the user and calculating the dynamic sensitivity of each sensitive region, the step comprises: selecting a time period to be judged, and obtaining a stay proportion index and an access stability index of the user in the candidate region based on the behavior characteristics of the user; wherein the stay proportion index is obtained based on the ratio of the number of trajectory points of the user in the candidate region in a preset time period to the total number of trajectory points in the candidate region , and the access stability index is obtained based on the access time of the user in the candidate region ; judging whether the stay proportion index and the access stability index meet a preset condition, if yes, determining that the candidate region is a sensitive region, obtaining the center position coordinates and the sensitivity score of the sensitive region, and constructing a sensitive region set.

[0014] According to an aspect of the present application, in the step of selecting a time period to be judged and obtaining a stay proportion index and an access stability index of the user in the candidate region based on the behavior characteristics of the user, the stay proportion index is expressed as: ; wherein, the stay proportion is represented by , the trajectory point is represented by , the candidate region is represented by , the total number of trajectory points in all candidate regions is represented by , and the indicator function is represented bya timestamp representing a trajectory point,,a start time and an end time representing a selected time period.a start time and an end time representing a selected time period.a start time and an end time representing a selected time period. According to an aspect of the present application, in the step of selecting a time period to be judged and obtaining a stay proportion indicator and an access stability indicator of a user in a candidate area based on the behavior characteristics of the user, the access stability indicator is obtained based on the following steps, which include: extracting a sequence of access times of trajectory points of a user in a candidate area within a preset time period, sorting the timestamps of all access times in the candidate area, and representing as: , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , a timestamp representing a trajectory point, , a start time and an end time representing a selected time period. a start time and an end time representing a selected time period.

[0015] According to an aspect of the present application, in the step of selecting a time period to be judged and obtaining a stay proportion indicator and an access stability indicator of a user in a candidate area based on the behavior characteristics of the user, the access stability indicator is obtained based on the following steps, which include: extracting a sequence of access times of trajectory points of a user in a candidate area within a preset time period, sorting the timestamps of all access times in the candidate area, and representing as: , ; wherein, represents a timestamp sequence of all access times of a user to a candidate area, , , , , represents a timestamp of the first, second, …, th visit to a candidate area; calculating the interval between adjacent visits based on the timestamp sequence of a user to a candidate area: ; wherein, represents the interval between adjacent visits, represents the time interval between the th and th visit, and , ; calculating the reciprocal of the interval standard deviation of the adjacent visit intervals to obtain the access stability indicator, wherein the access stability indicator is represented as: ; wherein, represents the access stability indicator.

[0016] According to an aspect of the present application, in the step of obtaining the central position coordinates and sensitivity scores of the sensitive areas and constructing a sensitive area set, it includes: obtaining the sensitivity scores of the sensitive areas, wherein the sensitivity score is represented as: ; wherein, represents the sensitivity score. The sensitive region set is constructed according to the obtained central position coordinates and sensitivity scores of the sensitive regions, and is represented as: ; ; ; wherein, represents the sensitive region set, , represents the central position coordinates of the sensitive region, represents the sensitivity score of the sensitive region, represents the sensitivity threshold, represents the longitude coordinate of the trajectory point , represents the latitude coordinate of the trajectory point .

[0017] According to an aspect of the present application, in step S2, in the step of constructing a three-dimensional dynamic noise model for generating a dynamically attenuated spatio-temporal noise field, the three-dimensional dynamic noise model is used to generate a dynamically attenuated spatio-temporal noise field in the spatial dimension, the temporal dimension, and the density dimension, wherein in the spatial dimension, the noise parameter is controlled to dynamically change based on the distance from the sensitive region center position, and if the distance from the sensitive region center position is farther, the noise parameter is controlled to exponentially attenuate, in the temporal dimension, the noise parameter is controlled to dynamically change based on the data time, and if the data time is farther from the current time, the noise parameter is controlled to exponentially attenuate, and in the density dimension, the noise parameter is controlled to dynamically change based on the regional population density, and if the regional population density decreases, the noise parameter is controlled to linearly attenuate.

[0018] According to an aspect of the present application, in step S2, in the step of constructing a three-dimensional dynamic noise model for generating a dynamically attenuated spatio-temporal noise field, the density dimension is taken as the basic noise term of the three-dimensional dynamic noise model, and the spatial dimension and the temporal dimension are taken as the attenuation terms of the three-dimensional dynamic noise model.

[0019] According to an aspect of the present application, in step S2, in the step of constructing a three-dimensional dynamic noise model for generating a dynamically attenuated spatio-temporal noise field, the three-dimensional dynamic noise model is represented as: ; ; ; ; wherein, is a basic minimum noise amplitude, is a population density scaling factor, is a current regional population density, is a reference population density, is a regional sensitivity, is a spatial attenuation coefficient, is a distance to a sensitive center, is a time attenuation amplitude, is a time attenuation rate, is a data timeliness, is a base attenuation coefficient, is a building density coefficient, is a regional building density, is a base attenuation rate, is a security level coefficient, is a data security level, is a current system time, is an original timestamp of the position data with perturbation.

[0020] According to one aspect of the present application, in step S3, in the step of generating a two-dimensional position offset based on the noise amplitude using a Laplace mechanism, the two-dimensional position offset is represented as: ; wherein, represents a longitude coordinate offset value in the two-dimensional position offset, represents a latitude coordinate offset value in the two-dimensional position offset, represents a Laplace mechanism, represents a noise amplitude generated by a three-dimensional dynamic noise model; In step S3, in the step of perturbing the position coordinates of the target position based on the two-dimensional position offset, the perturbed position coordinates are represented as: ; ; wherein, , represents a perturbed coordinate value, represents a timestamp of the trajectory point.

[0021] According to one aspect of the present application, the present application aims to establish a dynamic correlation mechanism between noise intensity and geographical position sensitivity, so that the noise amplitude can be adaptively adjusted to achieve collaborative optimization of strong protection in high-sensitivity areas and high precision in low-sensitivity areas.

[0022] According to one aspect of the present application, the present application proposes a sensitive area automatic discovery method based on trajectory behavior characteristics, which captures the spatio-temporal distribution of sensitive places in real time through data-driven algorithms, eliminates the dependence on manual configuration, and improves the dynamic adaptability of the system.

[0023] According to one scheme of the present application, a mathematical model of self-decay of noise intensity over time is designed, so that new data is fully protected while the analysis potential of historical data is gradually released, and the whole life cycle value management of data is realized.

[0024] According to one scheme of the present application, a new mechanism of dynamic adjustment of noise intensity according to geographic location, time and data value is designed, so that high-intensity protection of high-sensitive locations and high availability of data in low-sensitive areas are balanced under the premise of meeting the strict differential privacy requirements, and automatic sensitive area detection capability is constructed to eliminate the dependence on manual configuration and adapt to urban dynamic development.

[0025] According to one scheme of the present application, the balance between privacy protection and data utility is realized, original trajectory data is processed by a spatiotemporal density clustering algorithm, sensitive areas such as residential areas and hospitals are automatically identified, and accurate and comprehensive analysis of user behavior characteristics is realized based on the residence ratio and access stability, so that the area sensitivity score can be obtained without manual annotation of geographic fences.

[0026] According to one scheme of the present application, the sensitivity of each location is automatically identified according to the characteristics, and a spatiotemporal noise field is generated according to the sensitivity, so as to construct a dynamic noise model and realize the effect that the offset of high-sensitive areas is significantly greater than that of low-sensitive areas.

[0027] According to one scheme of the present application, high-sensitive areas are automatically identified by analyzing the residence ratio, access frequency and other behavior characteristics of users, and the dynamic sensitivity of each area is calculated, and the whole process does not need manual presetting of geographic fences or semantic labels, so that the original trajectory data can be converted into geographic block information with sensitivity marks.

[0028] According to one scheme of the present application, the fixed noise intensity leads to insufficient protection of high-sensitive areas such as residential areas and excessive distortion of low-sensitive areas such as commercial areas, and a dynamic decay noise field method is proposed, a three-dimensional dynamic noise model is constructed based on the output results of sensitive areas. In the spatial dimension, the noise intensity is adjusted according to the distance between the location and the sensitive center, and the closer the distance, the stronger the noise; in the time dimension, the noise is automatically decayed with the data timeliness, the protection intensity of new data is high and the disturbance of historical data is weak; in the density dimension, the noise amplitude is dynamically adjusted based on the population number of the area. The three are fused to generate a continuously distributed noise field function, which provides customized noise parameters for each location point.

[0029] According to one embodiment of the present invention, the core advantage of this invention compared to existing technologies lies in solving three major bottleneck problems in location data privacy protection: spatial indistinguishability, strong reliance on manual intervention, and conflicting timeliness. Firstly, through a sensitive area detection module, the nighttime activity patterns and access stability characteristics in user dwell patterns are analyzed, identifying highly sensitive areas without any manual annotation, significantly reducing the lag in protecting emerging sensitive points. Compared to the limitations of traditional Bayesian network solutions that rely on preset semantic labels, this method fully realizes data-driven dynamic perception, greatly reducing system maintenance costs and improving response speed.

[0030] According to one aspect of the present invention, regarding noise control mechanisms, this invention proposes a spatiotemporal noise field model, breaking through the rigid mode of traditional fixed noise. This model automatically adjusts the noise intensity based on the distance between the location and the sensitive center, causing high-intensity disturbances around residential areas while maintaining a slight shift in commercial areas, successfully achieving differentiated protection in the spatial dimension. Simultaneously, a time decay factor is introduced, ensuring sufficient privacy for new data while the noise intensity of historical data automatically decreases with time, effectively releasing the value of long-term data. This fundamentally avoids the problem of excessive distortion of historical data caused by neglecting timeliness in existing technologies. Attached Figure Description

[0031] Figure 1 This is a flowchart illustrating the steps of a differential privacy location protection method based on a dynamically attenuated noise field according to one embodiment of the present invention. Figure 2 This is a schematic diagram illustrating a scenario application of the differential privacy location protection method based on a dynamically attenuated noise field, according to one embodiment of the present invention. Detailed Implementation

[0032] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. The embodiments cannot be described in detail here, but the embodiments of the present invention are not limited to the following embodiments.

[0033] like Figure 1 As shown, according to one embodiment of the present invention, a differential privacy location protection method based on a dynamically attenuated noise field includes the following steps: S1. Obtain multiple candidate regions based on user trajectories, and identify different sensitive regions from the multiple candidate regions based on user behavior characteristics, and calculate the dynamic sensitivity of each sensitive region; wherein, user behavior characteristics include: user dwell time and access time in the candidate regions; S2. Construct a three-dimensional dynamic noise model for generating a dynamically decaying spatiotemporal noise field; wherein, the three-dimensional dynamic noise model is used to generate different noise amplitudes for different location coordinates and different timestamps in the sensitive region; S3. Obtain the position coordinates of the target position in the sensitive area and the corresponding time stamp, generate the corresponding noise amplitude value by using a three-dimensional dynamic noise model, generate a two-dimensional position offset based on the noise amplitude value by using a Laplace mechanism, perturb the position coordinates of the target position based on the two-dimensional position offset, and output the perturbed position coordinates.

[0034] According to an embodiment of the present application, in step S1, in the step of obtaining a plurality of candidate areas based on the user trajectory, the following steps are included: Collect a series of trajectory points in the user trajectory, and obtain a space-time feature vector of each trajectory point; in this embodiment, a corresponding fixed block is selected, and the user trajectory in the fixed block is collected based on the fixed block. The obtained user trajectory is discretely processed to obtain a series of continuous trajectory points; the series of trajectory points are represented by a trajectory point set, and are represented as follows: ; Wherein, represents the trajectory point set, represents the trajectory point, represents the longitude coordinate of the trajectory point , represents the latitude coordinate of the trajectory point , represents the time stamp of the trajectory point; Cluster the trajectory points based on the space-time feature vector and by using a clustering method to generate a plurality of candidate areas; in this embodiment, the clustering method used can be implemented by using an improved DBSCAN clustering algorithm, thereby realizing the effect of space-time density clustering on the original user trajectory data to automatically identify the corresponding candidate areas. In this embodiment, based on the space-time feature vector, the space-time clustering is performed, and the adjacent points within the distance radius and the time window are merged to form a candidate area , which is represented as: ; Wherein, represents the distance radius, and represent adjacent trajectory points, and represent the time stamps corresponding to the adjacent trajectory points and .

[0035] According to an embodiment of the present application, in step S1, in the step of identifying different sensitive areas from the plurality of candidate areas based on the behavior characteristics of the user and calculating the dynamic sensitivity of each sensitive area, the following steps are included: ​select a time period to be judged, and obtain a stay proportion index and an access stability index of the user in the candidate region based on the behavior characteristics of the user; the stay proportion index is obtained based on the ratio of the number of trajectory points of the user in the candidate region to the total number of trajectory points in the candidate region in a preset time period, and the access stability index is obtained based on the access time of the user in the candidate region ; In the embodiment, the stay proportion index is represented as: ; wherein, represents the stay proportion, represents the trajectory point, represents the candidate region, represents the total number of trajectory points in all candidate regions, represents an indicator function, , represents the start time and the end time of the selected time period.

[0036] Taking the night as an example, in order to analyze the night stay proportion of each candidate region, the start time and the end time of the corresponding time period can be set accordingly, for example, the start time of the night is set to 22:00 and the end time is set to 06:00 in the morning, then the stay proportion can be transformed into a relationship for the time period, that is: .

[0037] Further, the access stability index is obtained based on the following steps, which include: extracting the sequence of access times of the user trajectory points in the candidate region in a preset time period, and sorting the time stamps of all access times in the candidate region , and represented as: ; wherein, represents the time stamp sequence of all access times of the user to the candidate region , , , , represents the time stamp for the first, second, …, and time access to the candidate region ; calculate the adjacent access interval based on the time stamp sequence of the user to the candidate region : ; wherein, Indicates the interval between adjacent visits. Indicates the first Next and first The time interval between visits, and , ; The access stability index is obtained by calculating the reciprocal of the standard deviation of adjacent access intervals, where the access stability index is expressed as: ; in, This indicates an index of access stability.

[0038] The system determines whether the dwell rate and access stability metrics meet preset conditions. If they do, the candidate region is identified as a sensitive region. The center coordinates and sensitivity score of the sensitive region are then obtained, and a set of sensitive regions is constructed. (See [link to relevant documentation]). Figure 2 As shown. In this embodiment, the preset condition for determining the dwell rate indicator is a preset threshold. Therefore, by comparing the value of the dwell rate indicator with the set preset threshold, it can be determined whether the dwell rate indicator meets the standard for a sensitive area. Similarly, the preset condition for determining the access stability indicator is also a preset threshold. Therefore, by comparing the value of the access stability indicator with the set preset threshold, it can be determined whether the access stability indicator meets the standard for a sensitive area.

[0039] Through the above settings, this solution utilizes the designed dwell time ratio and access stability indicators to accurately quantify and analyze uncertain behavioral characteristics, greatly improving the accuracy of the analysis of sensitive areas. In particular, the set access stability indicator can more accurately and reliably filter out strong correlations between user behavior characteristics and candidate areas, thereby enabling more accurate identification of candidate areas.

[0040] According to one embodiment of the present invention, the step of obtaining the center coordinates and sensitivity score of a sensitive region and constructing a set of sensitive regions includes: Obtain the sensitivity score of the sensitive area, where the sensitivity score is expressed as: ; in, Indicates sensitivity score; Based on the sensitivity scoring set above, the dwell rate index and access stability index are effectively considered together. This allows for a more accurate and comprehensive evaluation of the sensitivity of corresponding areas based on relevant behavioral characteristics, which is more beneficial for improving the accuracy of sensitive area delineation.

[0041] The sensitive region set is constructed according to the obtained central position coordinates and sensitivity scores of the sensitive regions, and is represented as: ; ; ; wherein, represents the sensitive region set, 、 represents the central position coordinates of the sensitive region, represents the sensitivity score of the sensitive region, represents the sensitivity threshold, represents the longitude coordinate of the trajectory point , represents the latitude coordinate of the trajectory point .

[0042] In the embodiment, the sensitivity threshold can be set to 0.7.

[0043] According to an embodiment of the present application, in step S2, the step of constructing a three-dimensional dynamic noise model for generating a dynamically decaying spatio-temporal noise field and generating different noise field functions for different sensitive regions, the three-dimensional dynamic noise model is used to generate a dynamically decaying spatio-temporal noise field in the spatial dimension, the time dimension and the density dimension, wherein in the spatial dimension, the noise parameter is controlled to dynamically change based on the distance from the center position of the sensitive region, and if the distance from the center position of the sensitive region is far away, the noise parameter is controlled to exponentially decay, in the time dimension, the noise parameter is controlled to dynamically change based on the data time, and if the data time is farther away from the current time, the noise parameter is controlled to exponentially decay, and in the density dimension, the noise parameter is controlled to dynamically change based on the population density of the region, and if the population density of the region decreases, the noise parameter is controlled to linearly decay.

[0044] In the embodiment, the density dimension is used as the basic noise term of the three-dimensional dynamic noise model, and the spatial dimension and the time dimension are used as the decay terms of the three-dimensional dynamic noise model, that is, the overall configuration of the three-dimensional dynamic noise model is: basic noise term x spatial dimension decay term x time dimension decay term; wherein the priority of the spatial dimension decay term is higher than that of the time dimension decay term, so that the decay proportion of the position in the sensitive region is more prioritized, so as to realize more important privacy protection for the position.

[0045] According to an embodiment of the present application, in step S2, the step of constructing a three-dimensional dynamic noise model for generating a dynamically decaying spatio-temporal noise field, the three-dimensional dynamic noise model is represented as: ; ; ; ; wherein, is a base minimum noise amplitude, is a population density scaling factor, is a current area population density, is a reference population density, is an area sensitivity, is a spatial decay coefficient, is a distance to a sensitive center, is a time decay amplitude, is a time decay rate, is a data timeliness, is a base decay coefficient, is a building density coefficient, is an area building density, is a base decay rate, is a security level coefficient, is a data security level, is a current system time, is an original time stamp of the position data with perturbation.

[0046] According to an embodiment of the present application, in the step S3, the two-dimensional position offset is generated based on the noise amplitude using a Laplace mechanism, and the position coordinates of the target position are perturbed based on the two-dimensional position offset, and the perturbed position coordinates are taken as the output, and the two-dimensional position offset is expressed as: ; wherein, represents a longitude coordinate offset value in the two-dimensional position offset, represents a latitude coordinate offset value in the two-dimensional position offset, represents a Laplace mechanism, represents a noise amplitude generated by the three-dimensional dynamic noise model; Further, in the step S3, in the step of perturbing the position coordinates of the target position based on the two-dimensional position offset, the perturbed position coordinates are expressed as: ; ; wherein, , represents a perturbed coordinate value, represents a time stamp of the trajectory point.

[0047] According to the application, by analyzing the stay behavior in the position data, a high-sensitive area such as a residential area or a hospital is automatically identified and a dynamic score is calculated without manual annotation. The spatiotemporal noise field generation module constructs a dynamic noise model accordingly, generates customized noise parameters according to the three factors of spatial distance, time newness and population density. The noise parameters are adaptively called to generate a differential privacy protected offset for each position point, and a perturbed position is output. The whole process realizes automatic processing of sensitive area identification, noise dynamic modeling and position targeted perturbation, so that the residential area obtains strong offset protection and the commercial area maintains high precision, and the balance between privacy security and data utility is achieved.

[0048] The above is only an example of a specific solution of the present application, and for the devices and structures not described in detail, it should be understood that the general devices and general methods existing in the art are used to implement them.

[0049] The above is only one solution of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A differential privacy location protection method based on dynamically attenuated noise field, characterized in that, Includes the following steps: S1. Obtain multiple candidate regions based on user trajectories, and identify different sensitive regions from the multiple candidate regions based on user behavior characteristics, and calculate the dynamic sensitivity of each sensitive region; wherein, user behavior characteristics include: user dwell time and access time in the candidate regions; S2. Construct a three-dimensional dynamic noise model for generating a dynamically decaying spatiotemporal noise field; wherein, the three-dimensional dynamic noise model is used to generate different noise amplitudes for different location coordinates and different timestamps in the sensitive region; S3. Obtain the location coordinates and corresponding timestamp of the target location in the sensitive area and generate the corresponding noise amplitude using the three-dimensional dynamic noise model. Based on the noise amplitude, generate a two-dimensional position offset using the Laplace mechanism, and perturb the location coordinates of the target location based on the two-dimensional position offset, and output the perturbed location coordinates.

2. The differential privacy location protection method based on dynamically attenuated noise field according to claim 1, characterized in that, Step S1, which involves obtaining multiple candidate regions based on user trajectories, includes: Collect a series of trajectory points in the user trajectory and obtain the spatiotemporal feature vector of each trajectory point; Based on the spatiotemporal feature vector, the trajectory points are clustered using a clustering method to generate multiple candidate regions.

3. The differential privacy location protection method based on dynamically attenuated noise field according to claim 2, characterized in that, Step S1, which involves identifying different sensitive regions from multiple candidate regions based on user behavior characteristics and calculating the dynamic sensitivity of each sensitive region, includes: Select the time period to be judged, and obtain the user's dwell time ratio and access stability index in the candidate area based on the user's behavioral characteristics; wherein, the dwell time ratio is based on the user's behavior in the candidate area. The number of trajectory points and candidate regions within a preset time period The access stability index is obtained by the ratio of the number of all trajectory points in the candidate region. The time spent accessing the site; Determine whether the dwell rate index and the access stability index meet preset conditions. If they do, determine that the candidate area is a sensitive area, obtain the center coordinates and sensitivity score of the sensitive area, and construct a set of sensitive areas.

4. The differential privacy location protection method based on dynamically attenuated noise field according to claim 3, characterized in that, In the steps of selecting the time period to be judged and obtaining the user's dwell time ratio and access stability index in the candidate area based on the user's behavioral characteristics, the dwell time ratio index is expressed as: ； in, Indicates the percentage of stay. Represents the trajectory points, Indicates the candidate region. This represents the total number of trajectory points across all candidate regions. Indicates an indicator function, The timestamp representing the trajectory point , This indicates the start and end times of the selected time period.

5. The differential privacy location protection method based on dynamically attenuated noise field according to claim 4, characterized in that, In the steps of selecting the time period to be judged and obtaining the user's dwell time ratio and access stability index in the candidate area based on user behavior characteristics, the access stability index is obtained based on the following steps, which include: Extract candidate regions within a preset time period The sequence of access times for user trajectory points in the middle, and the candidate regions. Sort all access times by timestamps and represent them as follows: ； in, Indicates the user's preference for candidate regions A timestamp sequence of all access times. , , , Indicates the use of candidate regions The 1st, 2nd, ..., 1st The timestamp of the first visit; Based on user-selected candidate regions Calculate the adjacent access interval using timestamp sequences: ； in, Indicates the interval between adjacent visits. Indicates the first Next and first The time interval between visits, and , ; The access stability index is obtained by calculating the reciprocal of the standard deviation of the adjacent access intervals, wherein the access stability index is expressed as: ； in, This indicates an index of access stability.

6. The differential privacy location protection method based on dynamically attenuated noise field according to claim 5, characterized in that, The steps of obtaining the center coordinates and sensitivity scores of the sensitive regions and constructing a set of sensitive regions include: Obtain the sensitivity score of the sensitive region, wherein the sensitivity score is expressed as: ； in, Indicates sensitivity score; The set of sensitive regions is constructed based on the center coordinates and sensitivity scores of the acquired sensitive regions, and is represented as follows: ； ； ； in, Represents a set of sensitive regions. , Indicates the coordinates of the center location of the sensitive area. This indicates the sensitivity score of the sensitive area. Indicates the sensitivity threshold. Represents trajectory points longitude coordinates Represents trajectory points Latitude coordinates.

7. The differential privacy location protection method based on dynamically attenuated noise field according to claim 6, characterized in that, In step S2, the step of constructing a three-dimensional dynamic noise model for generating a dynamically decaying spatiotemporal noise field is described. This three-dimensional dynamic noise model is used to generate a dynamically decaying spatiotemporal noise field in spatial, temporal, and density dimensions. Specifically, in the spatial dimension, the noise parameters are dynamically changed based on the distance from the center of the sensitive area; if the distance is far from the center of the sensitive area, the noise parameters are controlled to decay exponentially. In the temporal dimension, the noise parameters are dynamically changed based on the timeliness of the data; if the data is older than the present, the noise parameters are controlled to decay exponentially. In the density dimension, the noise parameters are dynamically changed based on the regional population density; if the regional population density decreases, the noise parameters are controlled to decay linearly.

8. The differential privacy location protection method based on dynamically attenuated noise field according to claim 7, characterized in that, In step S2, the step of constructing a three-dimensional dynamic noise model for generating a dynamically decaying spatiotemporal noise field uses the density dimension as the basic noise term of the three-dimensional dynamic noise model, and the spatial dimension and the time dimension as the decay term of the three-dimensional dynamic noise model.

9. The differential privacy location protection method based on dynamically attenuated noise field according to claim 8, characterized in that, In step S2, the step of constructing a three-dimensional dynamic noise model for generating a dynamically decaying spatiotemporal noise field, wherein the three-dimensional dynamic noise model is represented as: ； ； ； ； in, Based on the minimum noise amplitude, Population density scaling factor Given the current population density of the area, For reference population density, For regional sensitivity, The spatial attenuation coefficient, Distance to the sensitive center The time decay magnitude, The decay rate over time. For data timeliness, Based on the attenuation coefficient, The building density coefficient, For regional building density, Based on the decay rate, For safety level coefficient, For data security level, The current system time. This is the original timestamp of the location data with perturbations.

10. The differential privacy location protection method based on dynamically attenuated noise field according to claim 9, characterized in that, In step S3, the step of generating a two-dimensional position offset using the Laplace mechanism based on the noise amplitude, the two-dimensional position offset is expressed as: ； in, This represents the longitude coordinate offset value in the two-dimensional position offset. This represents the latitude coordinate offset value in the two-dimensional position offset. Indicates the Laplace mechanism, This represents the noise amplitude generated by the three-dimensional dynamic noise model; In step S3, the perturbation of the target position coordinates based on the two-dimensional position offset is represented as follows: ； ； in, , This represents the coordinate values ​​after the disturbance. The timestamp representing the trajectory point.

Images