Track synthesis method, system and device based on space-time correlation and storage medium

Abstract

The invention relates to the technical field of information security, and relates to a trajectory synthesis method, system and device based on spatial-temporal correlation and a storage medium. The trajectory synthesis method based on spatio-temporal correlation comprises the following steps: acquiring a real trajectory data set authorized by a user; constructing a geographic sensing network according to the density distribution of the coordinate points in the real track data set, and generalizing the coordinate points into grids of the geographic sensing network; generating an initial starting and ending point relation matrix according to the starting point grid and the ending point grid of the track, and adding noise to elements in the initial starting and ending point relation matrix to obtain a noise-added starting and ending point relation matrix; extracting an initial transfer matrix of the trajectory under the plurality of time slices, and adding noise to elements in the initial transfer matrix to obtain a noise-added transfer matrix; and synthesizing a new trajectory by using the noise-added start-end point relation matrix and the noise-added transfer matrix to obtain a new trajectory data set. According to the new track data set synthesized by the method, the time information and the space information of the track are reserved, and the privacy information of the user is protected.

Description

Technical Field

[0001] The present invention relates to the field of information security technology, and particularly to a trajectory synthesis method, system, device and storage medium based on spatio-temporal correlation. Background Art

[0002] The spatio-temporal correlation of trajectories has important applications in many fields. When planning traffic roads, the correct spatio-temporal correlation of trajectories can be used to analyze the vehicle density map and judge the congestion conditions of different roads in a city at a certain moment, so as to better plan urban roads. For navigation software, the correct spatio-temporal correlation of trajectories can help it better plan driving routes for users and avoid congestion. The real trajectory dataset of vehicles has important value for traffic management and location services, but its release faces severe privacy challenges. Sensitive information (such as permanent address, commuting route, etc.) contained in the real trajectory dataset is easily reverse-inferred, and the release of the real trajectory dataset of vehicles may leak users' privacy information. Summary of the Invention

[0003] This application aims to at least solve the technical problems existing in the prior art, and provides a trajectory synthesis method, system, device and storage medium based on spatio-temporal correlation.

[0004] In a first aspect, the trajectory synthesis method based on spatio-temporal correlation provided by the present invention includes:[[]]

[0005] Obtain a real trajectory dataset, where the real trajectory dataset includes at least one trajectory authorized by a user, and each trajectory includes a plurality of coordinate points;

[0006] Construct a geographical perception network according to the density distribution of coordinate points in the real trajectory dataset, and generalize the position information of the coordinate points to the grids of the geographical perception network;

[0007] Construct an initial start-end point relationship matrix according to the position information corresponding to the start grid and the end grid of the trajectory in the geographical perception network, and add noise to the elements in the initial start-end point relationship matrix to obtain a noise-added start-end point relationship matrix;

[0008] Extract the initial transition matrix of the trajectory under multiple time slices, and add noise to the elements in the initial transition matrix to obtain a noise-added transition matrix;

[0009] Synthesize new trajectories by using the noise-added start-end point relationship matrix and the noise-added transition matrix to obtain a new trajectory dataset.

[0010] Optionally, the constructing an initial start-end point relationship matrix according to the position information corresponding to the start grid and the end grid of the trajectory in the geographical perception network, and adding noise to the elements in the initial start-end point relationship matrix according to a preset rule to obtain a noise-added start-end point relationship matrix includes:

[0011] Count the starting grid and the ending grid of the trajectories in the real trajectory dataset in the geosensory network, and generate an initial start-end relationship matrix;

[0012] Add Laplace noise to the elements of the initial start-end relationship matrix to obtain an intermediate start-end relationship matrix;

[0013] Normalize the elements of the intermediate start-end relationship matrix to obtain a final noisy start-end relationship matrix.

[0014] Optionally, the formula for adding Laplace noise to the elements of the initial start-end relationship matrix is:

[0015]

[0016] where C s represents the starting grid, C e represents the ending grid, s and e are both positive integers, and W represents the grid set of the geosensory network; U(C s , C e ) represents the number of trajectories with the starting grid C s and the ending grid C e ; Lap(1 / ε 1 ) represents the Laplace noise with the privacy budget ε 1 , ε 1 is the privacy budget parameter, represents the value after adding noise to the number of trajectories with the starting grid C s and the ending grid C e .

[0017] Optionally, the method for extracting the initial transition matrix of the trajectories under multiple time slices and adding noise to the elements in the initial transition matrix according to a preset rule to obtain a noisy transition matrix includes:

[0018] Slice the trajectories according to a time window to obtain multiple time slices;

[0019] Extract the grid transition pairs of the trajectories from time slice t to the next time slice t + 1. The grid transition pairs are used to represent the change information of the grid positions of the trajectories from time slice t to the next time slice t + 1, and t represents the index of the time slice;

[0020] Count the transition frequencies of the grid transition pairs within multiple time slices to obtain the transition probabilities between adjacent grids, and generate an initial transition matrix of the trajectories according to the transition probabilities between adjacent grids;

[0021] Add Laplace noise to the initial transition matrix to obtain a noisy transition matrix.

[0022] Optionally, the trajectory starts from grid C iTo grid C j The calculation formula for the transition probability is as follows:

[0023]

[0024] T(t,C i ,C j ) is the proportion of the number of grid transition pairs from grid C i to grid C j at time t to the number of grid transition pairs starting from C i at time t; represents the number of grid transition pairs starting from grid C i under time slice t. W t represents the grid transition pair under time slice t; "of consecutive locations C i ,C j inW t " represents the number of grid transition pairs from grid C t to grid C i in the grid transition pair W j .

[0025] Optionally, the method of synthesizing a new trajectory by using the noisy start-end relationship matrix and the noisy transition matrix to obtain a new trajectory dataset includes:

[0026] S61. Define the new trajectory dataset as an empty set;

[0027] S62. Randomly select the start grid and the end grid of the new trajectory to be synthesized from the noisy start-end relationship matrix;

[0028] S63. Starting from the start grid of the new trajectory to be synthesized, select the next grid of the new trajectory to be synthesized according to the transition probability data in the noisy transition matrix until the next grid of the new trajectory reaches the end grid of the new trajectory, obtaining the grid transition information of the new trajectory to be synthesized;

[0029] S64. Generate a new trajectory according to the grid transition information of the new trajectory to be synthesized and store the new trajectory in the new trajectory dataset;

[0030] Repeat steps S62 to S64 until the number of new trajectories in the new trajectory dataset is the same as the number of trajectories in the real trajectory dataset, obtaining the final new trajectory dataset.

[0031] Optionally, the step of generating a new trajectory according to the grid transition information includes:

[0032] Determine the grids passed by the new trajectory according to the grid transition information, and mark the grids passed by the new trajectory as target grids;

[0033] Analyze the geospatial perception network, extract the road information of the geospatial perception network, and mark the areas with roads in the geospatial perception network as reachable areas;

[0034] Randomly sample within the reachable area of the target grid to obtain the trajectory points corresponding to the target grid;

[0035] Connect the trajectory points according to the grid transfer order to obtain a new trajectory.

[0036] In a second aspect, the present invention provides a trajectory synthesis system based on spatio-temporal correlation, and the system includes:

[0037] An acquisition module, configured to acquire a real trajectory dataset, where the real trajectory dataset includes at least one trajectory authorized by a user, and each trajectory includes a plurality of coordinate points;

[0038] A geospatial perception network construction module, configured to construct a geospatial perception network according to the density distribution of the coordinate points in the real trajectory dataset, and generalize the position information of the coordinate points to the grids of the geospatial perception network;

[0039] A first processing module, configured to construct an initial start-end relationship matrix according to the position information of the start grid and the end grid corresponding to the trajectory in the geospatial perception network in the real trajectory dataset, and add noise to the elements in the initial start-end relationship matrix to obtain a noise-added start-end relationship matrix;

[0040] A second processing module, configured to extract the initial transition matrix of the trajectory under multiple time slices, and add noise to the elements in the initial transition matrix to obtain a noise-added transition matrix;

[0041] A trajectory synthesis module, configured to synthesize a new trajectory by using the noise-added start-end relationship matrix and the noise-added transition matrix to obtain a new trajectory dataset.

[0042] In a third aspect, the present invention provides an electronic device, and the electronic device includes:

[0043] At least one processor; and,

[0044] A memory communicatively connected to the at least one processor; wherein,

[0045] 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 so that the at least one processor can execute the above-mentioned trajectory synthesis method based on spatio-temporal correlation.

[0046] In a fourth aspect, the present invention further provides a computer-readable storage medium storing at least one computer program, and the at least one computer program is executed by a processor in an electronic device to implement the above-described trajectory synthesis method based on spatio-temporal correlation.

[0047] In summary, the present application includes the following beneficial technical effects:

[0048] By constructing a geographical perception network and mapping the coordinate points in the real trajectory dataset to the grid of the geographical perception network, the spatial information of the real trajectory can be effectively retained; by extracting the initial transition matrix of the trajectory under multiple time slices and using different initial transition matrices at different time points when synthesizing the trajectory, the time correlation of the trajectory can be effectively protected; the present application pays attention to both the spatial correlation and the time correlation of the trajectory data, making the synthesized trajectory dataset more authentic and more usable; by adding noise to the initial start-end relationship matrix and the initial transition matrix respectively and synthesizing new trajectories using the noisy start-end relationship matrix and the noisy transition matrix, the privacy information of users can be protected while retaining the time information and spatial information of the trajectory, and the data security is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic flowchart of a trajectory synthesis method based on spatio-temporal correlation provided by an embodiment of the present invention;

[0050] Figure 2 It is a schematic diagram of a trajectory synthesis algorithm provided by an embodiment of the present invention;

[0051] Figure 3 It is a schematic structural diagram of an electronic device for implementing the trajectory synthesis method based on spatio-temporal correlation provided by an embodiment of the present invention.

[0052] Reference numerals: 10, processor; 11, memory; 12, communication bus; 13, communication interface.

[0053] The realization, functional features and advantages of the objectives of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] The embodiments of the present invention will be described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation of the present invention.

[0055] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings. These are only for convenience in describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention.

[0056] In the description of the present invention, unless otherwise specified and defined, it should be noted that the terms "installation", "connection", and "coupling" should be understood in a broad sense. For example, it may be a mechanical connection or an electrical connection, or it may be the communication inside two elements. It can be directly connected or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.

[0057] Referring to Figure 1 as shown, it is a schematic flow chart of a trajectory synthesis method based on spatio-temporal correlation provided by an embodiment of the present invention. In this embodiment, the trajectory synthesis method based on spatio-temporal correlation includes:

[0058] S1. Obtain a real trajectory data set.

[0059] The real trajectory data set includes at least one vehicle trajectory authorized by the user. In this embodiment, the real trajectory data set includes n trajectories τ i , and the expression of the real trajectory data set is D = {τ 1 , τ 2 , …, τ i , …, τ n}, where n is the total number of trajectories in the real trajectory data set, i is a positive integer greater than 0 and less than or equal to n, and i is the index of the trajectory in the real trajectory data set; each trajectory contains l coordinate points in the format of (t, x, y), where t represents the time corresponding to the coordinate point, x represents the longitude of the coordinate point, and y represents the latitude of the coordinate point. The i-th trajectory

[0060] S2. Construct a geographical awareness network according to the density distribution of the coordinate points in the real trajectory data set, and generalize the position information of the coordinate points to the grids of the geographical awareness network.

[0061] The process of constructing the geographical awareness network is as follows:

[0062] First, the area where the trajectory is located needs to be divided into N grids of the same size. The range of the trajectory area can be determined according to the maximum and minimum values of the trajectory's longitude and latitude to determine the length and width of the constructed geographical perception network. Then, the grid size of the geographical perception network is further divided using the density of the trajectory coordinate points. In areas with a high density of coordinate points, the grid division is finer; in areas with a low density of coordinate points, the grid division is sparser. Finally, the geographical information of the vehicle driving area is obtained, and the divided grids are processed according to the real geographical network: grids without roads are set as inaccessible grids, grids with roads are set as accessible grids, the areas with roads in the accessible grids are set as accessible areas, and the areas without roads in the accessible grids are set as inaccessible areas. The accessibility between grids is processed in combination with the road distribution of the real geographical network to construct the geographical perception network. It should be added that the construction of the geographical perception network needs to meet the following constraints:

[0063] The movement of the trajectory points should be within the area where there are roads. The next coordinate point of the current trajectory coordinate point should be within the grid where the current coordinate point is located or in a grid adjacent to the grid where the current coordinate point is located. When a grid is accessible, the prerequisite is that there is a road connected to the current grid in it. Similarly, when converting the trajectory from grid form to coordinate point form later, the real geographical structure also needs to be referred to so that each point should be on a real and feasible road.

[0064] Through the geographical perception network, the accessibility relationship between a grid and the eight adjacent grids can be obtained, and an accessibility matrix is constructed based on the accessibility relationship. The accessibility matrix represents whether there is a path between grids through a matrix form. For example, if grid C 1 can reach grid C 2 through the accessible area, then the corresponding position from grid C 1 to grid C 2 in the accessibility matrix is 1. If grid C 1 cannot reach grid C 2 through the accessible area, then the corresponding position from grid C 1 to grid C 2 in the accessibility matrix is 0.

[0065] Generalize the points in the real trajectory dataset to the geographical perception network, and convert the original trajectory to τ i ’ = {cell 1 , cell 2 , …, cell l},

[0066] cell i represents the grid index of the geographical perception network, cell 1denote in the grid cell 1 the generalized trajectory dataset is denoted as D s The generalized trajectory dataset D s the trajectories in it will be composed in the form of a grid sequence

[0067] S3. Construct an initial start-end relationship matrix based on the position information of the start grid and the end grid corresponding to the trajectory in the geographical perception network, and add noise to the elements in the initial start-end relationship matrix to obtain a noisy start-end relationship matrix

[0068] trajectory τ i the start grid of is the grid position of the coordinate point corresponding to the start of trajectory τ i in the geographical perception network, and the end grid of trajectory τ i is the grid position of the coordinate point corresponding to the end of trajectory τ i in the geographical perception network

[0069] Specifically, constructing an initial start-end relationship matrix based on the position information of the start grid and the end grid corresponding to the trajectory in the geographical perception network, and adding noise to the elements in the initial start-end relationship matrix according to a preset rule to obtain a noisy start-end relationship matrix, including

[0070] S31. Count the start grid and the end grid of the trajectories in the real trajectory dataset in the geographical perception network, and generate an initial start-end relationship matrix

[0071] S32. Add Laplace noise to the elements of the initial start-end relationship matrix to obtain an intermediate start-end relationship matrix

[0072] The formula for adding Laplace noise to the elements of the initial start-end relationship matrix is

[0073]

[0074] where C s represents the start grid, C e represents the end grid, s and e are both positive integers, W represents the grid of the geographical perception network, that is, the set of all grids in the geographical perception network; U(C s , C e ) represents the number of trajectories with the start grid being C s and the end grid being C e ; Lap(1 / ε 1 ) represents the Laplace noise with the privacy budget being ε 1 , ε 1 is the privacy budget parameter in the differential privacy mechanism, which is used to control the noise intensity denotes for the start grid being C s, the end grid is C e The value after adding noise to the number of trajectories.

[0075] S33. Normalize the elements of the intermediate start-end relationship matrix to obtain the final noisy start-end relationship matrix.

[0076] The formula for normalizing the elements of the intermediate start-end relationship matrix is:

[0077]

[0078] Indicates that the trajectory starts from the start grid C s to the end grid C e The probability, and the values of s and e both range from [1, N], where N represents the number of grids in the geographical perception network; Indicates the trajectory dataset D s The number of pairs of start grids and end grids corresponding to all trajectories in, which is also equivalent to the number of trajectories in the trajectory dataset D s The number of trajectories.

[0079] Normalizing the elements of the intermediate start-end relationship matrix facilitates random sampling during subsequent synthesis of trajectories.

[0080] S4. Extract the initial transition matrix of the trajectories in the real trajectory dataset at multiple time slices, add noise to the elements in the initial transition matrix to obtain the noisy transition matrix.

[0081] Extract the initial transition matrix of the trajectories at multiple time slices, add noise to the elements in the initial transition matrix according to a preset rule to obtain the noisy transition matrix, including:

[0082] S41. Slice the trajectory according to the time window to obtain multiple time slices.

[0083] S42. Extract the grid transfer pairs of the trajectory from time slice t to the next time slice t + 1, and the grid transfer pairs are used to represent the change information of the grid position of the trajectory from time slice t to the next time slice t + 1.

[0084] S43. Statistically count the transfer frequencies of the grid transfer pairs in multiple time slices to obtain the transfer probabilities of adjacent grids in the geographical perception network, and generate the initial transition matrix of the trajectory according to the transfer probabilities of adjacent grids in the geographical perception network.

[0085] S44. Add Laplace noise to the initial transition matrix to obtain the noisy transition matrix.

[0086] In this embodiment, the trajectory dataset D s is sliced in time, and from each trajectory τ iExtract the transfer grid combinations at the same time point and the next time point, and form the first-order Markov matrix under each time slice through normalization.

[0087] Through the Markov matrices under multiple time slices, different Markov matrices are adopted at different time points when synthesizing trajectories, effectively protecting both the temporal correlation and spatial correlation of the trajectories.

[0088] From grid C i To grid C j The calculation formula for the transition probability is:

[0089]

[0090] t represents the index of the time slice, T(,C i ,C j ) is the proportion of the number of grid transition pairs from grid C i To grid C j At time t to the total number of grid transition pairs starting from C i At time t; Represents the number of grid transition pairs starting from grid C i At time slice t. W t Represents the grid transition pairs at time slice t; "of consecut i ve locat i ons C i ,C j in W t " represents the number of grid transition pairs from grid C t To grid C i To grid C j In the grid point pair W.

[0091] The calculation formula for adding Laplace noise to the value T(t,C i ,C j ) is:

[0092]

[0093] Lap(1 / ε 2 ) represents the Laplace noise with a privacy budget of ε 2 , Represents the result after adding noise to the value T(t,C i ,C j ), ε 2 Is the privacy budget parameter in the differential privacy mechanism, used to control the noise intensity.

[0094] In a preferred embodiment of the present embodiment, after obtaining the noise-added transition matrix, the trajectory synthesis method based on spatio-temporal correlation further includes: verifying the elements in the noise-added transition matrix by using a verification trajectory dataset to determine whether differential privacy is satisfied.

[0095] The specific proof process is as follows:

[0096] At a certain moment t, the transition matrix between grids is composed of those between all adjacent grids in the geo-aware grid network and is denoted as

[0097]

[0098] In this formula, n and m have no specific value references and are only used to represent unspecified adjacent grids; for example, in, grid C 1 and grid C n are adjacent grids, in grid C 1 and grid C m are adjacent grids, in, grid C m and grid C n are adjacent grids.

[0099] Since each T(t, C i , C j ) is perturbed by Lap(1 / ε 2 ), according to the Laplace mechanism, if the sensitivity Δf = 1, then satisfies ε 2 -differential privacy. The dataset and the dataset are used as verification datasets to verify the differential privacy of the noise-added transition matrix; where where τ represents a single trajectory, that is, the dataset has one more trajectory than the dataset ; we need to prove Δf represents the sensitivity; that is, there is:

[0100]

[0101] where, ‖.‖ represents the first norm, represents the encrypted value of T(t, C in the dataset i , C j ), represents the encrypted value of T(t, C in the dataset i , Cj )The encrypted numerical value after encryption.

[0102] Finally, since T(t, C i , C j ) for the entire trajectory is independent and non-overlapping, and T(t, C i , C j ) only depends on the current time node, it satisfies the parallel composition property. Therefore, for all time slices all satisfy ε 2 -differential privacy.

[0103]

[0104] Formula is used to normalize the Markov matrix (the noisy transition matrix) at time t for subsequent trajectory synthesis. At the same time, due to the post-processing property of differential privacy, the normalization operation conforms to ε 2 -differential privacy.

[0105] By adding noise to the initial transition matrix and the initial start-end point relationship matrix through differential privacy technology, the spatio-temporal correlation of the trajectory and the user's privacy information can be effectively protected, realizing efficient privacy protection for trajectory data.

[0106] S5. Synthesize new trajectories using the noisy start-end point relationship matrix and the noisy transition matrix to obtain a new trajectory dataset.

[0107] Compared with the existing solutions that only focus on one aspect of time correlation and space correlation, the trajectory dataset synthesized by the present invention is more realistic and has higher usability.

[0108] Specifically, synthesizing new trajectories using the noisy start-end point relationship matrix and the noisy transition matrix to obtain a new trajectory dataset includes:

[0109] S61. Define the new trajectory dataset as an empty set.

[0110] S62. Randomly select the start grid and the end grid of the new trajectory to be synthesized from the noisy start-end point relationship matrix.

[0111] S63. Starting from the start grid of the new trajectory to be synthesized, select the next grid of the new trajectory to be synthesized according to the transition probability data in the noisy transition matrix until the next grid of the new trajectory reaches the end grid of the new trajectory, obtaining the grid transition information of the new trajectory to be synthesized.

[0112] S64. Generate a new trajectory according to the grid transition information of the new trajectory to be synthesized and store the new trajectory in the new trajectory dataset.

[0113] Repeat steps S62 to S64 until the number of new trajectories in the new trajectory dataset is the same as the number of trajectories in the true trajectory dataset, obtaining the final new trajectory dataset.

[0114] In this embodiment, the noisy start and end point relationship matrix and the noisy transition matrix under multiple time slices are used to synthesize the new trajectory dataset D g The algorithm flow is as follows and can be referred to Figure 2 .

[0115] Specifically, after randomly sampling the start grid and end grid of the new trajectory from the noisy start and end point relationship matrix , the noisy start and end point relationship matrix and the noisy transition matrix are used to determine the length distribution of the trajectory In this embodiment, the length distribution of the trajectory can be constructed in the following manner

[0116]

[0117] where represents the possible length of the trajectory τ s with the starting position at C e and the ending position at C i . represents the normalized probability of passing through s C to e C through cells. It should be noted that is the shortest length (i.e., the minimum number of cells) from C to s C in e . For auxiliary calculation, we construct as a simple directed graph. Specifically, each cell is regarded as a node in the graph, and an edge is created between the cell and its movable cells, and the weight of each edge is set to 1. Using the well-known Dijkstra algorithm, we can calculate the between any two movable positions from this directed graph. By default, we also set the maximum length to where α is a preset safety factor, usually taken as 1.5 to limit the maximum length of the trajectory.

[0118] After determining the length of the trajectory , the next moving point is selected from the starting point according to the first-order Markov matrix at different time points and the end point direction until the end point is reached. Repeat this process until the new trajectory dataset D gThe number of trajectories in is the same as that in the real trajectory dataset D s The number of trajectories in is the same as that in the real trajectory dataset D

[0119] The trajectory dataset D generated using Algorithm 1 g The trajectory τ in i is still composed of transitions between grid cells i which need to be converted into trajectory points in the original dataset

[0120] In this embodiment, the steps of generating a new trajectory according to the grid transfer information include:

[0121] S641. Determine the grids passed by the new trajectory according to the grid transfer information, and mark the grids passed by the new trajectory as target grids;

[0122] S642. Analyze the geographical perception network, extract the road information of the geographical perception network, and mark the areas with roads in the geographical perception network as reachable areas;

[0123] S643. Randomly select coordinate points within the reachable area of the target grid to obtain the trajectory points corresponding to the target grid;

[0124] For each trajectory point, it should be sampled from the range of its corresponding grid. When sampling, the range of random sampling needs to be restricted in combination with the real geographical network to ensure that the sampled points are on the real roads. Point sampling is performed within the reachable area of the target grid, and the grid points are converted into longitude and latitude coordinates and time coordinate points to obtain the trajectory points corresponding to the target grid.

[0125] S644. Connect the trajectory points in the grid transfer order to obtain a new trajectory.

[0126] Based on the same inventive concept, an embodiment of the present invention provides a trajectory synthesis system based on spatio-temporal correlation.

[0127] The trajectory synthesis system based on spatio-temporal correlation according to the present invention can be installed in an electronic device. According to the functions to be realized, the trajectory synthesis system based on spatio-temporal correlation includes an acquisition module, a geographical perception network construction module, a first processing module, a second processing module, and a trajectory synthesis module,

[0128] The acquisition module can acquire a real trajectory dataset, and the real trajectory dataset includes at least one trajectory authorized by the user, and each trajectory includes a plurality of coordinate points;

[0129] The geographical perception network construction module can construct a geographical perception network according to the density distribution of the coordinate points in the real trajectory dataset, and generalize the position information of the coordinate points to the grids of the geographical perception network;

[0130] The first processing module can construct an initial start-end point relationship matrix based on the position information corresponding to the start grid and the end grid of the trajectory in the geographical perception network in the real trajectory dataset, and add noise to the elements in the initial start-end point relationship matrix to obtain a noisy start-end point relationship matrix;

[0131] The second processing module can extract the initial transition matrix of the trajectory under multiple time slices, and add noise to the elements in the initial transition matrix to obtain a noisy transition matrix;

[0132] The trajectory synthesis module can synthesize new trajectories using the noisy start-end point relationship matrix and the noisy transition matrix to obtain a new trajectory dataset.

[0133] The module described in the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device.

[0134] The various change modes and specific examples in the trajectory synthesis method based on spatio-temporal correlation provided in the above embodiments are equally applicable to the trajectory synthesis system based on spatio-temporal correlation in this embodiment. Through the foregoing detailed description of the trajectory synthesis method based on spatio-temporal correlation, those skilled in the art can clearly know the implementation method of the trajectory synthesis system based on spatio-temporal correlation in this embodiment. For the sake of brevity of the specification, it will not be elaborated here.

[0135] This application also discloses an electronic device, as Figure 3 shown, which is a schematic structural diagram of the electronic device of the trajectory synthesis method based on spatio-temporal correlation provided by an embodiment of the present invention. The electronic device may include at least one processor 10, a memory 11 communicatively connected to at least one processor, a communication bus 12, and a communication interface 13, and may further include a computer program stored in the memory 11 and executable on the processor 10, such as a trajectory synthesis method program based on spatio-temporal correlation.

[0136] Among them, the processor 10 can be composed of integrated circuits in some embodiments. For example, it can be composed of a single packaged integrated circuit, or can be composed of multiple packaged integrated circuits with the same or different functions, including a combination of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control core (Control Unit) of the electronic device, connecting various components of the entire electronic device through various interfaces and circuits, running or executing programs or modules stored in the memory 11 (such as executing a trajectory synthesis method based on spatio-temporal correlation, etc.), and calling data stored in the memory 11 to perform various functions of the electronic device and process data.

[0137] The memory 11 includes at least one type of readable storage medium. The readable storage medium includes flash memory, mobile hard disks, multimedia cards, card-type memories (such as SD or DX memories, etc.), magnetic memories, magnetic disks, optical discs, etc. The memory 11 can be an internal storage unit of the electronic device in some embodiments, such as the 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 also include both the internal storage unit and the external storage device of the electronic device. The memory 11 can not only be used to store application software installed on the electronic device and various types of data, such as the code of a trajectory synthesis method program based on spatio-temporal correlation, etc., but also be used to temporarily store data that has been output or will be output.

[0138] The communication bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is set to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0139] The communication interface 13 is used for communication between the above-mentioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), and is generally used to establish a communication connection between this electronic device and other electronic devices. The user interface may be a display, an input unit (such as a keyboard), and optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch liquid crystal display, and an OLED (Organic Light-Emitting Diode) toucher, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, and is used to display the information processed in the electronic device and to display a visual user interface.

[0140] Figure 3 Only the electronic device with components is shown. Those skilled in the art can understand that Figure 3 The shown structure does not constitute a limitation on the electronic device, and it may include fewer or more components than shown, or combine certain components, or have different component arrangements. For example, although not shown, the electronic device may further include a power source (such as a battery) for powering each component. Preferably, the power source can be logically connected to at least one processor 10 through a power management device, so as to implement functions such as charging management, discharging management, and power consumption management through the power management device. The power source may also include any components such as one or more DC or AC power sources, a recharge device, a power failure detection circuit, a power converter or an inverter, and a power status indicator. The electronic device may also include a variety of sensors, a Bluetooth module, a Wi-Fi module, etc., which will not be elaborated here.

[0141] It should be understood that the embodiments are only for illustration purposes and are not limited by this structure in the scope of the patent application.

[0142] Furthermore, if the integrated module / unit of the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium may be volatile or non-volatile.

[0143] The embodiments of the present application provide a computer-readable storage medium, for example, including: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM, Read-Only Memory). This computer-readable storage medium stores a computer program that can be loaded and executed by a processor to perform the trajectory synthesis method based on spatio-temporal correlation in the above embodiments.

[0144] In the description of this specification, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", "a kind of implementation manner", "a preferred implementation manner" or "some examples", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0145] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and purposes of the present invention, and the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A trajectory synthesis method based on spatiotemporal correlation, characterized in that: The method comprises: Acquire a real trajectory data set, where the real trajectory data set includes at least one trajectory authorized by a user, and each trajectory includes a number of coordinate points; A geo-aware network is constructed based on the density distribution of coordinate points in the real trajectory dataset, and the location information of the coordinate points is generalized to the grid of the geo-aware network; According to the location information of the starting point grid and the ending point grid corresponding to the trajectory in the geographic perception network in the real trajectory data set, an initial start-end point relationship matrix is ​​constructed, and noise is added to the elements in the initial start-end point relationship matrix to obtain a noisy start-end point relationship matrix; Extract the initial transfer matrix of the trajectory under multiple time slices, add noise to the elements in the initial transfer matrix, and obtain the noisy transfer matrix; The new trajectory is synthesized using the noisy start-end relationship matrix and the noisy transfer matrix to obtain a new trajectory dataset.

2. The trajectory synthesis method based on spatiotemporal correlation according to claim 1, characterized in that: The method constructs an initial start-end relationship matrix according to the position information corresponding to the start grid and the end grid of the trajectory in the geographic perception network, and adds noise to the elements in the initial start-end relationship matrix according to a preset rule to obtain a noisy start-end relationship matrix, including: Count the starting point grids and ending point grids of the trajectories in the real trajectory dataset in the geographic perception network to generate the initial starting and ending point relationship matrix; Add Laplace noise to the elements of the initial start-end relationship matrix to obtain the intermediate start-end relationship matrix; The elements of the intermediate start-end relationship matrix are normalized to obtain the final noisy start-end relationship matrix.

3. The trajectory synthesis method based on spatiotemporal correlation according to claim 2, characterized in that: The formula for adding Laplace noise to the elements of the initial start-end relationship matrix is: Among them, C s represents the starting grid, C e represents the end point grid, s and e are both positive integers, and W represents the grid set of the geographic awareness network; U(C s ,C e ) indicates that the starting grid is C s , the end grid is C e The number of trajectories; Lap(1 / ε1) represents the Laplace noise with a privacy budget of ε1, and ε1 is the privacy budget parameter; Indicates that the starting grid is C s , the end grid is C e The number of trajectories after adding noise.

4. The trajectory synthesis method based on spatiotemporal correlation according to claim 1, characterized in that: The initial transfer matrix of the extracted trajectory under multiple time slices, adding noise to the elements in the initial transfer matrix according to a preset rule to obtain a noisy transfer matrix, includes: Slice the trajectory according to the time window to obtain multiple time slices; Extract the grid transfer pair of the trajectory from time slice t to the next time slice t+1. The grid transfer pair is used to represent the change information of the grid position when the trajectory moves from time slice t to the next time slice t+1. t represents the index of the time slice. Count the transfer frequencies of grid transfer pairs in multiple time slices to obtain the transfer probabilities of adjacent grids, and generate the initial transfer matrix of the trajectory according to the transfer probabilities of adjacent grids; Laplace noise is added to the initial transfer matrix to obtain the noisy transfer matrix.

5. The trajectory synthesis method based on spatiotemporal correlation according to claim 4, characterized in that: Trajectory from grid C i To Grid C j The calculation formula of the transition probability is: T(t,C i ,C j ) is the time from grid C at time t i To Grid C j The number of grid transfer pairs at time t is C i is the ratio of the number of grid transfer pairs to the starting point; Represents the time slice t with a grid C i is the number of grid transfer pairs from the starting point. t represents the grid transfer pair under time slice t; "of consecutive locations C i ,C j in W t ” indicates the effect of grid transfer on W t From the grid C i To Grid C j The number of mesh transfer pairs.

6. The trajectory synthesis method based on spatiotemporal correlation according to claim 4, characterized in that: The method of synthesizing a new trajectory using a noisy start-end relationship matrix and a noisy transfer matrix to obtain a new trajectory data set includes: S61, defining the new trajectory data set as an empty set; S62, randomly selecting a starting point grid and an ending point grid of a new trajectory to be synthesized from the noisy start-end relationship matrix; S63, starting from the starting grid of the new trajectory to be synthesized, selecting the next grid of the new trajectory to be synthesized according to the transition probability data in the noisy transfer matrix, until the next grid of the new trajectory reaches the end grid of the new trajectory, and obtaining the grid transfer information of the new trajectory to be synthesized; S64, generating a new trajectory according to the grid transfer information of the new trajectory to be synthesized, and storing the new trajectory in a new trajectory data set; Steps S62 to S64 are repeatedly performed until the number of new trajectories in the new trajectory dataset is consistent with the number of trajectories in the real trajectory dataset, thereby obtaining a final new trajectory dataset.

7. The trajectory synthesis method based on spatiotemporal correlation according to claim 1, characterized in that: The step of generating a new trajectory according to the grid transfer information of the new trajectory to be synthesized includes: Determine the grid that the new trajectory passes through according to the grid transfer information, and mark the grid that the new trajectory passes through as the target grid; Parse the geo-aware network, extract the road information of the geo-aware network, and mark the areas with roads in the geo-aware network as reachable areas; Randomly sample within the reachable area of ​​the target grid to obtain the trajectory points corresponding to the target grid; Connect the trajectory points according to the grid transfer order to obtain a new trajectory.

8. A trajectory synthesis system based on time-space correlation, used to implement the trajectory synthesis method based on time-space correlation according to any one of claims 1 to 7, characterized in that: include: An acquisition module is used to acquire a real trajectory data set, where the real trajectory data set includes at least one trajectory authorized by the user, and each trajectory includes a number of coordinate points; The geo-aware network construction module is used to construct a geo-aware network according to the density distribution of coordinate points in the real trajectory dataset, and generalize the location information of the coordinate points to the grid of the geo-aware network; The first processing module is used to construct an initial start-end relationship matrix according to the position information corresponding to the start grid and the end grid of the trajectory in the geographic perception network in the real trajectory data set, and add noise to the elements in the initial start-end relationship matrix to obtain a noisy start-end relationship matrix; The second processing module is used to extract the initial transfer matrix of the trajectory under multiple time slices, add noise to the elements in the initial transfer matrix, and obtain the noisy transfer matrix; The trajectory synthesis module is used to synthesize a new trajectory using the noisy start-end relationship matrix and the noisy transfer matrix to obtain a new trajectory data set.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor (10); and, a memory (11) communicatively connected to the at least one processor (10); The memory (11) stores a computer program executable by the at least one processor (10), and the computer program is executed by the at least one processor (10) so that the at least one processor (10) can execute the trajectory synthesis method based on spatiotemporal correlation as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program; when the computer program is executed by a processor, the trajectory synthesis method based on spatiotemporal correlation as claimed in any one of claims 1 to 7 is implemented.

Images