A mobile signaling trajectory data generation method and system based on diffusion process

Abstract

The application provides a mobile signaling trajectory data generation method and system based on a diffusion process, and the steps of the method include: obtaining original mobile signaling data of a preset time period, the original mobile signaling data including mobile signaling data corresponding to each base station, and preprocessing the original mobile signaling data; determining a base station relationship graph based on the mobile signaling data of each base station, the base station relationship graph including the switching relationship between base stations, and constructing an initial data set of each base station based on the switching relationship of each base station and the characteristic data of the base station; constructing a representation vector corresponding to each base station based on the initial data set of each base station, combining the representation vectors of the base stations into an input vector, inputting the input vector into a pre-trained diffusion model, the diffusion model outputting a trajectory generation vector, the trajectory generation vector being composed of a plurality of trajectory generation sub-vectors corresponding to users; and determining the trajectory data of the users in the preset time period based on the trajectory generation vector.

Description

Technical Field

[0001] This invention relates to the field of trajectory generation technology, and in particular to a method and system for generating mobile signaling trajectory data based on a diffusion process. Background Technology

[0002] The generation of trajectory data relies on various sensors, algorithms, and application scenarios, making it a complex and precise process. Real trajectory data is primarily obtained by sampling the movement of a moving object within a spatiotemporal environment. Each sampling point contains key information such as the moving object's position, timestamp, and velocity. Connecting these sampling points according to the sampling order constitutes the trajectory data.

[0003] However, existing trajectory generation technologies are often designed for continuous trajectory data generated by positioning devices such as GPS. GPS-based trajectory data requires user authorization, has limited coverage, and restricts its application scope. Mobile signaling data, collected by mobile operators, typically features a large user base, wide coverage, accuracy, and real-time performance, generating significant application value in areas such as population monitoring, public safety, and intelligent transportation. However, due to concerns about personal privacy and data security risks, the availability of accurate individual mobile trajectory data based on mobile signaling remains significantly limited in practical applications. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method for generating mobile signaling trajectory data based on diffusion processes, in order to eliminate or improve one or more defects existing in the prior art.

[0005] One aspect of the present invention provides a method for generating mobile signaling trajectory data based on a diffusion process, the method comprising the following steps:

[0006] The raw mobile signaling data for a preset time period is obtained. The raw mobile signaling data includes the mobile signaling data corresponding to each base station. The raw mobile signaling data is then preprocessed.

[0007] A base station relationship diagram is determined based on the mobile signaling data of each base station. The base station relationship diagram includes the handover relationship between base stations. An initial dataset for each base station is constructed based on the handover relationship of each base station and the characteristic data of each base station.

[0008] Based on the initial dataset of each base station, a representation vector corresponding to each base station is constructed. The representation vectors of each base station are combined into an input vector and input into a pre-trained diffusion model. The diffusion model outputs a trajectory generation vector, which is composed of multiple trajectory generation sub-vectors corresponding to multiple users.

[0009] The trajectory data of the user within a preset time period is determined based on the trajectory generation vector.

[0010] Using the above scheme, this scheme takes the mobile signaling data of the base station as the basis. First, the mobile signaling data is preprocessed to delete duplicate data and other invalid data. Then, a base station relationship diagram is constructed, which includes the handover relationship between base stations. The handover relationship between base stations is the relationship between the user switching from the connection of one base station to the connection of another base station. A representation vector corresponding to each base station is constructed. Then, the trajectory generation vector is output through the diffusion model to obtain the user's trajectory data. This scheme can obtain coarse-grained user movement trajectory based on the mobile signaling data of the base station.

[0011] In some embodiments of the present invention, in the step of preprocessing the raw mobile signaling data:

[0012] The original mobile signaling data within a preset time period is divided into multiple sub-time periods of mobile signaling data;

[0013] The base station to which the user is connected is determined based on the mobile signaling data for each sub-time period.

[0014] In some embodiments of the present invention, in the step of determining the base station connected to by a user in a sub-time period based on the mobile signaling data of each sub-time period, the base station with the longest connection time for each user in the sub-time period is counted as the base station connected to by the user in the sub-time period.

[0015] In some embodiments of the present invention, the preprocessing step of preprocessing the raw mobile signaling data includes sorting by time, removing ping-pong data, removing duplicate data, and removing noisy data.

[0016] In some embodiments of the present invention, in the step of determining the base station relationship graph based on the mobile signaling data of each base station, each base station is treated as a node in the base station relationship graph, and if there is a handover relationship between two base stations, an edge is constructed between the two base stations.

[0017] In some embodiments of the present invention, the feature data of the base station includes information such as the base station's location area identification code, cell identification code, and latitude and longitude. In the step of constructing a representation vector corresponding to each base station based on the initial dataset of each base station:

[0018] The feature data of the base stations in the initial dataset are combined with the handover relationship of each base station and then encoded.

[0019] The representation vector for each base station is obtained.

[0020] In some embodiments of the present invention, the value of each dimension in the trajectory generation vector corresponds to a base station. In the step of determining the user's trajectory data in a preset time period based on the trajectory generation vector, the user's trajectory data for the base station is determined based on the dimensional order in the trajectory generation vector.

[0021] In some embodiments of the present invention, the method further includes pre-training the diffusion model, wherein the pre-training step includes:

[0022] The predicted noise loss, recovery loss, and base station handover constraint loss are calculated based on the output trajectory generation vector and label data.

[0023] The total loss function is calculated based on the predicted noise loss, recovery loss, and base station handover constraint loss, and the diffusion model is pre-trained based on the total loss function.

[0024] In some embodiments of the present invention, in the step of calculating the prediction noise loss, recovery loss, and base station handover constraint loss based on the output trajectory generation vector and label data:

[0025] The mean square error between the predicted noise and white noise is calculated as the predicted noise loss.

[0026] The mean square error between the output trajectory generation vector and the input vector is calculated and used as the recovery loss;

[0027] Traverse adjacent base stations in the trajectory generation vector. If two adjacent base stations have an edge in the base station relationship graph, the loss value is 0; if two adjacent base stations do not have an edge in the base station relationship graph, the loss value is 1. Accumulate the loss values ​​to obtain the base station handover constraint loss.

[0028] A second aspect of the present invention also provides a mobile signaling trajectory data generation system based on a diffusion process. The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor executes the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method described above.

[0029] A third aspect of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned mobile signaling trajectory data generation method based on diffusion process.

[0030] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the text, or may be learned by practice of the invention. The objects and other advantages of the invention will become apparent from the description and the accompanying drawings.

[0031] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0032] The accompanying drawings, which are provided to further illustrate the invention and form part of this application, are not intended to limit the scope of the invention.

[0033] Figure 1 This is a schematic diagram of one embodiment of the mobile signaling trajectory data generation method based on diffusion process of the present invention;

[0034] Figure 2 This is a schematic diagram of the processing architecture of the mobile signaling trajectory data generation method based on the diffusion process of the present invention. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0036] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0037] In practical implementation, mobile signaling data typically possesses characteristics such as a large user base, wide coverage, authenticity, objectivity, and strong real-time performance, generating significant application value in fields such as population monitoring, public safety, and intelligent transportation. However, due to concerns about personal privacy and data security risks, the availability of authentic individual mobile trajectory data based on mobile signaling remains highly limited in practical applications. Furthermore, due to the instability of communication, storage, and device performance, raw mobile signaling trajectory data may contain substantial redundancy, missing data, and noise, severely impacting data usability. Against this backdrop, constructing a mobile signaling trajectory data generation model to produce synthetic data that is statistically close to and can replace real data at the application level is one effective solution to these problems. However, most existing technologies only address GPS trajectory data; currently, there is no technology for generating trajectory data specifically for mobile signaling. To address these issues, this invention proposes a method for generating trajectory data based on mobile signaling.

[0038] like Figure 1 and 2 As shown, this invention proposes a method for generating mobile signaling trajectory data based on a diffusion process. The steps of this method include:

[0039] Step S100: Obtain the original mobile signaling data for a preset time period. The original mobile signaling data includes the mobile signaling data corresponding to each base station. The original mobile signaling data is then preprocessed.

[0040] In practice, mobile signaling data is the data generated during communication between mobile phone users and base stations. It mainly includes information such as user location, call start and stop, and SMS sending and receiving. This data has wide application value in many fields such as network planning, user behavior analysis, and traffic flow monitoring.

[0041] Step S200: Determine a base station relationship diagram based on the mobile signaling data of each base station. The base station relationship diagram includes the handover relationship between base stations. Construct an initial dataset for each base station based on the handover relationship of each base station and the feature data of each base station.

[0042] In the specific implementation process, the characteristic data of the base station includes information such as the base station's location area code (LAC), cell identification code (CI), and latitude and longitude.

[0043] Step S300: Construct a representation vector corresponding to each base station based on the initial dataset of each base station, combine the representation vectors of each base station into an input vector and input it into the pre-trained diffusion model. The diffusion model outputs a trajectory generation vector, which is composed of multiple trajectory generation sub-vectors corresponding to the user.

[0044] In practical implementation, the diffusion model includes a forward noise addition module and a reverse noise reduction module:

[0045] Forward noise addition module: Uses a diffusion model to gradually add noise to the trajectory representation vector to obtain a noise vector with added noise;

[0046] Inverse denoising module: First, based on the noise vector obtained by the forward diffusion denoising module, the noise is predicted by the denoising network. Then, the initial noise vector is recovered based on the predicted noise and the noise vector obtained by the forward diffusion module.

[0047] In the specific implementation process, given any time step, a predicted noise is generated based on the pre-trained denoising network. Then, the cosine similarity between the predicted noise and the base station representation vector is calculated, and the generated trajectory vector is obtained using the argmax operator.

[0048] The argmax operator is a widely used concept in mathematical optimization, statistical analysis, and machine learning. Its core function is to find and return the value or index of the independent variable that maximizes a given function. When dealing with a series of numerical or function values, the argmax operator precisely locates the maximum value by comparing these values. In the context of arrays or lists, the argmax operator returns the index of the largest element, which is crucial for fast data retrieval and analysis. In function optimization scenarios, the argmax operator helps researchers pinpoint the parameter combinations that lead to the extrema of the objective function, thereby driving decision-making and model optimization. It is worth noting that the argmax operator does not directly provide the maximum value itself, but rather indicates the location of the maximum value; this characteristic makes it irreplaceable in data science and machine learning.

[0049] In the specific implementation process, during the processing steps of the forward noise-adding module:

[0050] First, set the initial values ​​for the model parameters, such as the diffusion step T and the noise β. t , where β t It follows a standard normal distribution with a mean of 0 and a variance of 1.

[0051] Let x0 be the initial input (i.e., input vector) of a trajectory T. A diffusion step t is randomly sampled from 1 to T, and noise is added to x0. The noise data x is generated according to the following formula. t :

[0052]

[0053] Where ∈ tIt is white noise sampled from a standard normal distribution N(0,I), where I is an identity matrix with the same dimensions as the input x0, and α t =1-β t

[0054]

[0055] In the processing steps of the reverse noise reduction module:

[0056] From noise data x T Initially, noise is removed gradually over T time steps. At each step t, the noise ε is predicted by the denoising network DN. θ (x t ,t), that is: ε θ (x t ,t)=DN θ (x T ), where θ is a learnable parameter. This invention uses a sequence generation-related model architecture such as Diffwave as the denoising network.

[0057] Based on the predicted noise ε θ (x t (t) and current noise data x t Try to gradually restore x t-1 Let the recovered x t-1 Represented as It can be obtained from the following formula:

[0058]

[0059] Let the recovered x0 be...

[0060] Calculated recovery The cosine similarity between the base station representation vector and the generated base station trajectory is obtained as a sequence using argmax.

[0061] Step S400: Determine the user's trajectory data within a preset time period based on the trajectory generation vector.

[0062] Using the above scheme, this scheme takes the mobile signaling data of the base station as the basis. First, the mobile signaling data is preprocessed to delete duplicate data and other invalid data. Then, a base station relationship diagram is constructed, which includes the handover relationship between base stations. The handover relationship between base stations is the relationship between the user switching from the connection of one base station to the connection of another base station. A representation vector corresponding to each base station is constructed. Then, the trajectory generation vector is output through the diffusion model to obtain the user's trajectory data. This scheme can obtain coarse-grained user movement trajectory based on the mobile signaling data of the base station.

[0063] In some embodiments of the present invention, in the step of preprocessing the raw mobile signaling data:

[0064] The original mobile signaling data within a preset time period is divided into multiple sub-time periods of mobile signaling data;

[0065] The base station to which the user is connected is determined based on the mobile signaling data for each sub-time period.

[0066] In some embodiments of the present invention, in the step of determining the base station connected to by a user in a sub-time period based on the mobile signaling data of each sub-time period, the base station with the longest connection time for each user in the sub-time period is counted as the base station connected to by the user in the sub-time period.

[0067] In some embodiments of the present invention, the time of a preset time period is divided into H time periods, where H = 288, that is, each time period is a sub-time period of 5 minutes, and the base station where the user stays for the longest time in 5 minutes is extracted as the location within that time period.

[0068] In some embodiments of the present invention, the preprocessing step of preprocessing the raw mobile signaling data includes sorting by time, removing ping-pong data, removing duplicate data, and removing noisy data.

[0069] In some embodiments of this invention, ping-pong data refers to data generated in mobile communication networks due to frequent switching between two or more base stations caused by changes in base station signal strength. This type of data superficially indicates a change in the user's location, but in reality, the user's location has not moved at all. Therefore, it interferes with data analysis and needs to be identified and removed.

[0070] The method for removing ping-pong data typically involves the following steps:

[0071] Data preprocessing: The collected signaling data is preprocessed, including data cleaning and format conversion, to ensure the accuracy and consistency of the data.

[0072] Identifying the ping-pong effect: By analyzing base station handover records in signaling data, data segments exhibiting the ping-pong effect can be identified. This is typically achieved by calculating the time interval and frequency of handovers between adjacent base stations. If the handover frequency is too high (e.g., exceeding a certain threshold), the ping-pong effect is considered to exist.

[0073] Removing Ping-Pong Data: For identified ping-pong data segments, you can choose to remove them directly or perform smoothing. Direct removal means deleting the data segments exhibiting the ping-pong effect from the dataset; smoothing means using algorithms to correct the ping-pong data to obtain data that more closely resembles the user's actual movement trajectory.

[0074] In some embodiments of the present invention, in the step of determining the base station relationship graph based on the mobile signaling data of each base station, each base station is treated as a node in the base station relationship graph, and if there is a handover relationship between two base stations, an edge is constructed between the two base stations.

[0075] In practical implementation, the base station relationship diagram can be represented using an adjacency matrix. Specifically, the topological relationship can be represented by an adjacency matrix A. In the adjacency matrix, if two base stations have an edge, the value is set to 1 at the corresponding position of the two base stations; otherwise, it is 0.

[0076] In some embodiments of the present invention, the feature data of the base station includes information such as the base station's location area identification code, cell identification code, and latitude and longitude. In the step of constructing a representation vector corresponding to each base station based on the initial dataset of each base station:

[0077] The feature data of the base stations in the initial dataset are combined with the handover relationship of each base station and then encoded.

[0078] The representation vector for each base station is obtained.

[0079] In the specific implementation process, the location area identification code (LAC) of the base station is represented by one-hot encoding. The dimension of the base station's feature data can be the number of different LACs, or other information such as the base station's latitude and longitude address can be added as a supplement to the base station's features.

[0080] In the specific implementation process, in the step of combining and encoding the feature data of the base stations in the initial dataset with the handover relationship of each base station, a preset encoder is used for encoding. Specifically, the encoder can be a graph autoencoder, a graph variational autoencoder, or a graph attention autoencoder.

[0081] In the specific implementation process, unsupervised learning graph neural networks such as graph autoencoders (GAE), graph variational autoencoders (VGAE), and graph attention autoencoders (GATE) are used to learn the vector of each base station, transforming the base station into a continuous representation vector. Where d is the dimension of the feature.

[0082] In some embodiments of the present invention, the value of each dimension in the trajectory generation vector corresponds to a base station. In the step of determining the user's trajectory data in a preset time period based on the trajectory generation vector, the user's trajectory data for the base station is determined based on the dimensional order in the trajectory generation vector.

[0083] In some embodiments of the present invention, the method further includes pre-training the diffusion model, wherein the pre-training step includes:

[0084] The predicted noise loss, recovery loss, and base station handover constraint loss are calculated based on the output trajectory generation vector and label data.

[0085] The total loss function is calculated based on the predicted noise loss, recovery loss, and base station handover constraint loss, and the diffusion model is pre-trained based on the total loss function.

[0086] In the specific implementation process, in the step of calculating the total loss function based on the predicted noise loss, recovery loss and base station handover constraint loss, the sum of the predicted noise loss, recovery loss and base station handover constraint loss is calculated as the total loss function.

[0087] In some embodiments of the present invention, in the step of calculating the prediction noise loss, recovery loss, and base station handover constraint loss based on the output trajectory generation vector and label data:

[0088] The mean square error between the predicted noise and white noise is calculated as the predicted noise loss.

[0089] The mean square error between the output trajectory generation vector and the input vector is calculated and used as the recovery loss;

[0090] Traverse adjacent base stations in the trajectory generation vector. If two adjacent base stations have an edge in the base station relationship graph, the loss value is 0; if two adjacent base stations do not have an edge in the base station relationship graph, the loss value is 1. Accumulate the loss values ​​to obtain the base station handover constraint loss.

[0091] In the specific implementation process, due to the spatial relationship between base stations during handover, in order to ensure that adjacent base stations in the generated trajectory data still retain the original handover spatial relationship, base station handover spatial relationship constraints are added. The model loss is divided into three parts: prediction noise loss. Recovering losses and base station handover constraint loss

[0092] in For the predicted noise ε θ (x t ,t) and white noise ∈ t The mean square error between them is expressed as:

[0093]

[0094] For recovery The mean square error between the original x0 and the mean square error is expressed as:

[0095]

[0096] For the generated base station trajectory sequence, traverse adjacent base stations. and If two adjacent and If there are no edges in the original base station relationship graph G, meaning there is no handover relationship, then the loss value is 1; otherwise, the loss value is 0. The sum of all loss values ​​is then calculated.

[0097] The loss function of the final model is

[0098] In some embodiments of the present invention, during the processing of the diffusion model:

[0099] First, generate noise that conforms to a standard normal distribution. Given a diffusion step T, generate predicted noise ε based on the already trained network DN. T Then generate according to the following formula To noise

[0100]

[0101] Calculate prediction noise Similarly, the cosine similarity to the base station representation vector is used to obtain the generated base station trajectory sequence using argmax.

[0102] This invention also provides a mobile signaling trajectory data generation system based on a diffusion process. The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor executes the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method described above.

[0103] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned method for generating mobile signaling trajectory data based on a diffusion process. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0104] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.

[0105] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0106] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0107] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for mobile signaling trajectory data generation based on diffusion process, characterized in that, The steps of the method include: Obtaining original mobile signaling data of a preset time period, the original mobile signaling data including mobile signaling data corresponding to each base station, and preprocessing the original mobile signaling data; Determining a base station relationship graph based on the mobile signaling data of each base station, the base station relationship graph including switching relationships between base stations, constructing an initial data set of each base station based on the switching relationship of each base station and the characteristic data of the base station; Constructing a representation vector corresponding to each base station based on the initial data set of each base station, combining the representation vectors of the base stations into an input vector, and inputting the input vector into a pre-trained diffusion model, the diffusion model outputting a trajectory generation vector, the trajectory generation vector being composed of a plurality of trajectory generation sub-vectors corresponding to users; Determining trajectory data of the users in the preset time period based on the trajectory generation vector.

2. The mobile signaling trajectory data generation method based on diffusion process according to claim 1, wherein, In the step of preprocessing the original mobile signaling data: Divide the original mobile signaling data of the preset time period into mobile signaling data of a plurality of sub-time periods; Determine the base station connected by the user in each sub-time period based on the mobile signaling data of each sub-time period.

3. The mobile signaling trajectory data generation method based on diffusion process according to claim 2, wherein, In the step of determining the base station connected by the user in each sub-time period based on the mobile signaling data of each sub-time period, count the base station with the longest connection duration of each user in the sub-time period as the base station connected by the user in the sub-time period.

4. The mobile signaling trajectory data generation method based on diffusion process according to claim 1, wherein, In the step of preprocessing the original mobile signaling data, the preprocessing includes time sorting, removing ping-pong data, removing duplicate data, and removing noise data.

5. The mobile signaling trajectory data generation method based on diffusion process according to claim 1, wherein, In the step of determining the base station relationship graph based on the mobile signaling data of each base station, each base station is taken as a node in the base station relationship graph, and if there is a switching relationship between two base stations, an edge between the two base stations is constructed.

6. The mobile signaling trajectory data generation method based on diffusion process according to claim 1, wherein, The characteristic data of the base station includes location area identification code, cell identification code, and longitude and latitude information of the base station, and in the step of constructing a representation vector corresponding to each base station based on the initial data set of each base station: Combine the characteristic data of the base station in the initial data set with the switching relationship of each base station, and encode them; Obtain the representation vector corresponding to each base station.

7. The mobile signaling trajectory data generation method based on diffusion process according to claim 1, wherein, Each dimension in the trajectory generation vector corresponds to a base station, and in the step of determining the trajectory data of the users in the preset time period based on the trajectory generation vector, the trajectory data of the users for the base stations is determined based on the dimension order in the trajectory generation vector.

8. The mobile signaling trajectory data generation method based on diffusion process according to any one of claims 1 to 7, characterized in that, The steps of the method further include pre-training the diffusion model, and in the pre-training step: Calculate the prediction noise loss, the recovery loss, and the base station switching constraint loss based on the output trajectory generation vector and the label data; Calculate the total loss function based on the prediction noise loss, the recovery loss, and the base station switching constraint loss, and pre-train the diffusion model based on the total loss function.

9. The mobile signaling trajectory data generation method based on diffusion process according to claim 8, wherein, In the step of calculating the prediction noise loss, the recovery loss, and the base station switching constraint loss based on the output trajectory generation vector and the label data: Calculate the mean square error between the predicted noise and the white noise as the prediction noise loss; Calculate the mean square error between the output trajectory generation vector and the input vector as the recovery loss; The loss value is accumulated to obtain a base station switching constraint loss.

10. A mobile signaling trajectory data generation system based on diffusion processes, characterized in that, The system comprises a computer device, the computer device comprises a processor and a memory, the memory stores computer instructions, and the processor is configured to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method as claimed in any one of claims 1-9.

Images