Motion trajectory analysis method, device, equipment and storage medium

Abstract

The present disclosure provides a motion trajectory analysis method and device, equipment and storage medium, which relates to the technical field of communication and can accurately analyze the motion trajectory of a terminal based on signaling data. The method comprises: obtaining signaling data of at least one terminal in a target area, the signaling data recording the connection of the terminal with a base station in mobile communication, so that the position of the terminal can be analyzed based on the position of the base station. The base station model established based on the antenna directional diagram of all base stations in the target area can be used to represent the comprehensive signal strength at different positions in the target area. The connection between the terminal and the base station is affected by the signal strength of the base station at the position of the terminal, so that the motion trajectory analysis based on convolution of the base station model and the signaling data of the terminal can correct the connection position of the terminal with the base station in the signaling data based on the signal strength of the base station, thereby improving the accuracy of the motion trajectory analysis of the terminal based on the signaling data.

Description

Technical Field

[0001] This disclosure relates to the field of communication technology, and in particular to a motion trajectory analysis method, apparatus, device, and storage medium. Background Technology

[0002] With the widespread adoption of smart mobile devices and the development of wireless communication technology, massive amounts of user spatiotemporal trajectory data have been generated. By analyzing this data, we can discover user movement patterns and further uncover patterns of population clustering. For example, analyzing user spatiotemporal trajectory information can enrich user profiles, predict user behavior, and reveal potential user needs. In addition, it plays an important role in commercial promotion such as precise personalized marketing and urban governance.

[0003] Currently, the Global Positioning System (GPS) is commonly used for positioning and motion trajectory analysis. Although GPS positioning can achieve accurate positioning by receiving GPS satellite signals through the GPS chip built into the terminal, GPS signal loss often occurs when there are buildings blocking the view or when the user does not continuously turn on the GPS positioning function of the terminal, making it impossible to guarantee the determination of an accurate spatiotemporal trajectory. Summary of the Invention

[0004] This disclosure provides a motion trajectory analysis method, apparatus, device, and storage medium, which are used to provide a method that can accurately determine spatiotemporal trajectories.

[0005] To achieve the above objectives, the present disclosure adopts the following technical solution:

[0006] Firstly, a motion trajectory analysis method is provided, the method comprising:

[0007] Acquire signaling data of terminals located within the target area; wherein the aforementioned signaling data is used to record the base stations connected to by the aforementioned terminals;

[0008] Based on the antenna patterns of all base stations within the target area, a base station model for the target area is established; the base station model is used to indicate the overall signal strength within the target area.

[0009] Based on the aforementioned base station model and the signaling data of the aforementioned terminal, convolution-based motion trajectory analysis is performed to obtain the motion trajectory information of the aforementioned terminal.

[0010] In conjunction with the first aspect above, in one possible implementation, the establishment of a base station model for the target area based on the antenna patterns of all base stations within the target area includes:

[0011] Obtain the antenna pattern of all base stations within the target area mentioned above;

[0012] Perform Fourier transform on the antenna pattern of each base station to obtain the antenna pattern of each base station in Fourier space.

[0013] The antenna patterns of each base station in Fourier space are linearly superimposed to obtain the base station model described above.

[0014] In conjunction with the first aspect above, in one possible implementation, the motion trajectory information includes the spatiotemporal trajectory of the terminal; the convolution-based motion trajectory analysis based on the base station model and the signaling data of the terminal to obtain the motion trajectory information of the terminal includes:

[0015] The terminal location matrix of the aforementioned terminal is determined based on the aforementioned signaling data; the terminal location matrix of the aforementioned terminal is a set of geographical locations of the base stations that the aforementioned terminal has connected to.

[0016] The above base station model is used as a convolution kernel to convolve the above terminal location matrix to determine the approximate spatiotemporal trajectory of the above terminal;

[0017] The estimated spatiotemporal trajectory is determined based on the aforementioned approximate spatiotemporal trajectory and the road location matrix within the aforementioned target area; wherein the aforementioned road location matrix is ​​the set of location coordinates of all roads within the aforementioned target area;

[0018] Gaussian blurring is applied to the estimated spatiotemporal trajectory and the road position matrix to obtain the Gaussian blurred estimated spatiotemporal trajectory and the Gaussian blurred road position matrix respectively.

[0019] Calculate the product of the estimated spatiotemporal trajectory after Gaussian blurring and the road position matrix after Gaussian blurring, and determine the spatiotemporal trajectory of the terminal based on the product.

[0020] In conjunction with the first aspect above, in one possible implementation, the signaling data also records the connection duration when the terminal connects to the base station; the motion trajectory information includes stopping points;

[0021] The above-mentioned convolution-based motion trajectory analysis, based on the base station model and the signaling data of the terminal, is used to obtain the motion trajectory information of the terminal, including:

[0022] Based on the aforementioned signaling data, the terminal location matrix of the aforementioned terminal and the connection duration matrix between the aforementioned terminal and the base station are determined; wherein, the terminal location matrix of the aforementioned terminal is a set of geographical locations of the base stations that the aforementioned terminal has connected to, and the connection duration matrix is ​​a set of connection durations between the aforementioned terminal and the base stations that it has connected to.

[0023] The above base station model is used as a convolution kernel to convolve the above connection duration matrix and the above terminal location matrix to determine the time-weighted approximate spatiotemporal trajectory of the above terminal based on the connection duration; the above time-weighted approximate spatiotemporal trajectory is the set of connection durations of the above terminal with the base station at each location;

[0024] Based on the predetermined dwell time threshold and the aforementioned time-weighted approximate spatiotemporal trajectory, the dwell point of the terminal is determined.

[0025] In conjunction with the first aspect above, in one possible implementation, determining the terminal's stopping point based on a predetermined stopping time threshold and the aforementioned time-weighted approximate spatiotemporal trajectory includes:

[0026] Determine whether the connection duration corresponding to the first position point in the above time-weighted approximate spatiotemporal trajectory is greater than the above dwell time threshold. If it is greater, then determine the above first position point as a candidate dwell point.

[0027] Using the candidate stopping points as the search center, search for points of interest within a preset stopping distance threshold range. The points of interest are locations that are pre-marked with geographical attributes.

[0028] If no point of interest exists, the candidate resting point is determined as the resting point of the terminal; if a point of interest exists, the point of interest whose distance from the candidate resting point is less than or equal to a preset distance threshold is determined as the resting point of the terminal.

[0029] In conjunction with the first aspect above, in one possible implementation, when the aforementioned stopping point is a work / residence location, the aforementioned connection duration matrix is ​​a set of connection durations within a preset time range from the connection durations between the terminal and the base station determined based on the aforementioned signaling data.

[0030] In conjunction with the first aspect above, in one possible implementation, the terminal is included in multiple terminals, and the method further includes:

[0031] Based on the signaling data of the aforementioned multiple terminals, the cluster location matrix corresponding to the aforementioned multiple terminals and the cluster connection duration matrix connecting the aforementioned multiple terminals to the base station are determined; wherein the cluster location matrix corresponding to the aforementioned multiple terminals is a linear superposition of the terminal location matrix of each of the aforementioned terminals, the terminal location matrix of the aforementioned terminals is a set of geographical locations of the base stations connected to the aforementioned terminals, and the cluster connection duration matrix is ​​a linear superposition of the normalized connection duration matrices connecting the aforementioned terminals to the base station.

[0032] The above base station model is used as a convolution kernel to convolve the cluster connection duration matrix of the above multiple terminals and the base station and the above cluster location matrix to determine the cluster approximate spatiotemporal trajectory of the above multiple terminals; the above cluster approximate spatiotemporal trajectory is the set of normalized connection durations of the above multiple terminals and the base station at each location.

[0033] Based on the predetermined aggregation threshold and the aforementioned cluster approximate spatiotemporal trajectory, it is determined whether the second position point in the aforementioned cluster approximate spatiotemporal trajectory is an aggregation point, and the aforementioned second position is located on the aforementioned cluster approximate spatiotemporal trajectory.

[0034] Secondly, a motion trajectory analysis device is provided, the device comprising:

[0035] The signaling data acquisition module is used to acquire signaling data of terminals within the target area; wherein the aforementioned signaling data is used to record the base stations that the aforementioned terminals have connected to;

[0036] The base station model building module is used to build a base station model of the target area based on the antenna patterns of all base stations in the target area; the base station model is used to indicate the overall signal strength in the target area.

[0037] The motion trajectory information analysis module is used to perform convolution-based motion trajectory analysis based on the aforementioned base station model and the signaling data of the aforementioned terminal to obtain the motion trajectory information of the aforementioned terminal.

[0038] In conjunction with the second aspect above, in one possible implementation, the base station model establishment module includes:

[0039] Antenna pattern acquisition unit, used to acquire antenna patterns of all base stations in the target area mentioned above;

[0040] The Fourier transform unit is used to perform Fourier transform on the antenna pattern of each base station to obtain the antenna pattern of each base station in Fourier space.

[0041] The base station model building unit is used to linearly superimpose the antenna patterns of each base station in Fourier space to obtain the base station model mentioned above.

[0042] In conjunction with the second aspect above, in one possible implementation, the motion trajectory information includes the spatiotemporal trajectory of the terminal; the motion trajectory information analysis module includes:

[0043] The terminal location matrix determination unit is used to determine the terminal location matrix of the terminal based on the aforementioned signaling data; the terminal location matrix of the terminal is a set of geographical locations of the base stations that the terminal has connected to.

[0044] The approximate spatiotemporal trajectory determination unit is used to convolve the above-mentioned base station model as a convolution kernel onto the above-mentioned terminal position matrix to determine the approximate spatiotemporal trajectory of the above-mentioned terminal.

[0045] The estimated spatiotemporal trajectory determination unit is used to determine the estimated spatiotemporal trajectory based on the aforementioned approximate spatiotemporal trajectory and the road position matrix within the aforementioned target area; wherein the aforementioned road position matrix is ​​a set of position coordinates of all roads within the aforementioned target area;

[0046] Gaussian blur unit is used to perform Gaussian blur on the estimated spatiotemporal trajectory and the road position matrix respectively, so as to obtain the Gaussian blurred estimated spatiotemporal trajectory and the Gaussian blurred road position matrix.

[0047] The spatiotemporal trajectory determination unit is used to calculate the product of the estimated spatiotemporal trajectory after Gaussian blurring and the road position matrix after Gaussian blurring, and to determine the spatiotemporal trajectory of the terminal based on the product.

[0048] In conjunction with the second aspect above, in one possible implementation, the signaling data also records the connection duration when the terminal connects to the base station; the motion trajectory information includes stopping points;

[0049] The aforementioned motion trajectory information analysis module includes:

[0050] The terminal location and duration determination unit is used to determine the terminal location matrix of the terminal and the connection duration matrix between the terminal and the base station based on the signaling data; wherein, the terminal location matrix of the terminal is a set of geographical locations of the base stations that the terminal has connected to, and the connection duration matrix is ​​a set of connection durations between the terminal and the base stations that it has connected to.

[0051] The time-weighted approximate spatiotemporal trajectory determination unit is used to convolve the above-mentioned base station model as a convolution kernel to the above-mentioned connection duration matrix and the above-mentioned terminal location matrix to determine the time-weighted approximate spatiotemporal trajectory of the above-mentioned terminal based on the connection duration; the above-mentioned time-weighted approximate spatiotemporal trajectory is the set of connection durations of the above-mentioned terminal connecting with the base station at each location;

[0052] The stopping point determination unit is used to determine the stopping point of the terminal based on a pre-determined stopping time threshold and the aforementioned time-weighted approximate spatiotemporal trajectory.

[0053] In conjunction with the second aspect above, in one possible implementation, the aforementioned stopping point determination unit includes:

[0054] The candidate stopping point determination subunit is used to determine whether the connection duration corresponding to the first position point in the above time-weighted approximate spatiotemporal trajectory is greater than the above stopping time threshold. If it is greater, the above first position point is determined as a candidate stopping point.

[0055] The point of interest search subunit is used to search for points of interest within a preset stopping distance threshold range with the above candidate stopping points as the search center. The above points of interest are locations that are pre-marked with geographical attributes.

[0056] The stopping point determination subunit is used to determine the candidate stopping point as the stopping point of the terminal if no point of interest exists; if a point of interest exists, the point of interest whose distance from the candidate stopping point is less than or equal to a preset distance threshold is determined as the stopping point of the terminal.

[0057] In conjunction with the second aspect above, in one possible implementation, when the aforementioned stopping point is a work-residence location, the aforementioned connection duration matrix is ​​a set of connection durations within a preset time range from the connection durations between the terminal and the base station determined based on the aforementioned signaling data.

[0058] In conjunction with the second aspect above, in one possible implementation, the terminal is included among multiple terminals, and the device further includes:

[0059] The cluster location and duration determination module is used to determine the cluster location matrix corresponding to the above-mentioned multiple terminals and the cluster connection duration matrix between the above-mentioned multiple terminals and the base station based on the signaling data of the above-mentioned multiple terminals; wherein the cluster location matrix corresponding to the above-mentioned multiple terminals is a linear superposition of the terminal location matrix of each terminal among the multiple terminals, the terminal location matrix of the above-mentioned terminals is a set of geographical locations of the base stations connected to by the above-mentioned terminals, and the cluster connection duration matrix is ​​a linear superposition of the normalized connection duration matrices between each terminal among the multiple terminals and the base station;

[0060] The cluster approximate spatiotemporal trajectory determination module is used to convolve the above base station model as a convolution kernel to the cluster connection duration matrix and the cluster location matrix of the above multiple terminals to determine the cluster approximate spatiotemporal trajectory of the above multiple terminals; the above cluster approximate spatiotemporal trajectory is the set of normalized connection durations of the above multiple terminals connecting with the base station at each location.

[0061] The aggregation point determination module is used to determine whether a second position point in the aforementioned cluster approximate spatiotemporal trajectory is an aggregation point based on a pre-determined aggregation threshold and the aforementioned cluster approximate spatiotemporal trajectory, wherein the aforementioned second position point is located on the aforementioned cluster approximate spatiotemporal trajectory.

[0062] Thirdly, a motion trajectory analysis device is provided, comprising: a processor and a memory; wherein the memory is used to store computer execution instructions, and when the motion trajectory analysis device is running, the processor executes the computer execution instructions stored in the memory to cause the motion trajectory analysis device to perform the motion trajectory analysis method as described in the first aspect and any possible implementation thereof.

[0063] Fourthly, this disclosure provides a computer-readable storage medium storing instructions that, when executed by a processor of a motion trajectory analysis device, enable the motion trajectory analysis device to perform the motion trajectory analysis method as described in the first aspect and any possible implementation thereof.

[0064] In this disclosure, the name of the aforementioned motion trajectory analysis device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear under other names. As long as the function of each device or functional module is similar to that of this disclosure, it falls within the scope of the claims of this disclosure and its equivalents.

[0065] These or other aspects of this disclosure will become more readily apparent in the following description.

[0066] The technical solution provided in this disclosure offers at least the following beneficial effects: This application provides a motion trajectory analysis method. This method acquires signaling data of a terminal within a target area, wherein the signaling data is used to record the base stations connected to by the terminal. Based on the antenna patterns of all base stations within the target area, a base station model of the target area is established. This base station model is used to indicate the overall signal strength within the target area. Based on the base station model and the terminal's signaling data, convolution-based motion trajectory analysis is performed to obtain the terminal's motion trajectory information. Since the base station model indicates the overall signal strength within the target area, by performing convolution-based motion trajectory analysis on the base station model and the terminal's signaling data, noise in the signaling data can be eliminated through convolution, and the features in the signaling data can be enhanced, thereby accurately obtaining the motion trajectory. Attached Figure Description

[0067] Figure 1 This is a schematic diagram of the hardware structure of a motion trajectory analysis device provided in this disclosure;

[0068] Figure 2 This is a flowchart illustrating a motion trajectory analysis method provided in this disclosure;

[0069] Figure 3 A flowchart illustrating another motion trajectory analysis method provided in this disclosure;

[0070] Figure 4 A flowchart illustrating another motion trajectory analysis method provided in this disclosure;

[0071] Figure 5 A flowchart illustrating another motion trajectory analysis method provided in this disclosure;

[0072] Figure 6 This is a schematic diagram of the structure of a motion trajectory analysis device provided in this disclosure. Detailed Implementation

[0073] The motion trajectory analysis method, apparatus, device, and storage medium provided in the embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0074] The terms “first” and “second” in this disclosure and its accompanying drawings are used to distinguish different objects or to distinguish different treatments of the same object, rather than to describe a particular order of objects.

[0075] Furthermore, the terms “comprising” and “having”, and any variations thereof, used in the description of this disclosure are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus.

[0076] It should be noted that in this disclosure, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" or "for example" in this disclosure should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0077] Figure 1 This is a schematic diagram of the structure of a motion trajectory analysis device provided in an embodiment of this disclosure. Figure 1 As shown, the motion trajectory analysis device 100 includes at least one processor 101, a communication line 102, and at least one communication interface 104, and may also include a memory 103. The processor 101, memory 103, and communication interface 104 are connected via the communication line 102.

[0078] The processor 101 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this disclosure, such as one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs).

[0079] Communication line 102 may include a path for transmitting information between the aforementioned components.

[0080] The communication interface 104 is used to communicate with other devices or communication networks. It can use any transceiver-like device, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), etc.

[0081] The memory 103 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of including or storing desired program code having the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0082] In one possible design, the memory 103 can exist independently of the processor 101, meaning the memory 103 can be an external memory of the processor 101. In this case, the memory 103 can be connected to the processor 101 via a communication line 102 to store execution instructions or application code, and its execution is controlled by the processor 101 to implement the motion trajectory analysis method provided in the following embodiments of this disclosure. In another possible design, the memory 103 can also be integrated with the processor 101, meaning the memory 103 can be an internal memory of the processor 101. For example, the memory 103 can be a cache, used to temporarily store some data and instruction information.

[0083] As one possible implementation, processor 101 may include one or more CPUs, for example Figure 1 CPU0 and CPU1 in the example. Alternatively, the motion trajectory analysis device 100 may include multiple processors, such as CPU0 and CPU1. Figure 1 The processors 101 and 107 are included. Alternatively, the motion trajectory analysis device 100 may also include an output device 105 and an input device 106.

[0084] Through the above description of the implementation methods, those skilled in the art will clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the network node can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, modules, and network nodes described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0085] The following is an explanation of the terms used in this application.

[0086] 1. Fourier Transform

[0087] The Fourier transform is a linear integral transform used to transform signals between the time (or spatial) domain and the frequency domain. In different applications, the Fourier transform has different forms, such as the continuous Fourier transform and the discrete Fourier transform. The discrete Fourier transform uses discretized data, facilitating fast computation by computers. The two-dimensional discrete Fourier transform is represented as:

[0088]

[0089] in This indicates the Fourier transform operation. This represents the inverse Fourier transform operation, j is the imaginary unit, (x, y) represents two orthogonal directions in the real space plane, and N... x N y Let (u,v) represent the number of grid cells in the x and y directions of the discrete data, respectively, and (u,v) are the two orthogonal directions corresponding to (x,y) in the Fourier space.

[0090] 2. Fast Fourier Transform (FFT)

[0091] The Fast Fourier Transform (FFT) is a fast algorithm for the Discrete Fourier Transform (DFT). The time complexity of directly calculating the DFT is O(n log n). 2 The time complexity of the Fast Fourier Transform (FFT) to obtain the same result is O(n log n), where n is the number of cells in the discrete data. The larger the value of n, the more significant the computational savings of the FFT.

[0092] 3. Convolution Theorem

[0093] The convolution theorem states that the Fourier transform of the convolution of two functions is the product of the Fourier transforms of the two functions:

[0094]

[0095] By applying the convolution theorem and combining it with the fast Fourier transform, the time required to compute convolutions can be greatly reduced, achieving the goal of fast computation.

[0096] 4. Plane approximation

[0097] In reality, the Earth's surface is a sphere. However, since base station operations are typically city-level (e.g., a city-level branch of a telecom operator), the area of ​​business operations is generally limited to a single city. A typical city spans approximately 100 kilometers, the Earth's average radius is 6371 kilometers, and the angular size at the center of the sphere is θ = 100 / 6371 = 0.0156961 radians, satisfying sinθ ~ θ. Therefore, although the Earth's surface is curved, it can be approximated as a plane within the span of a typical city. The systematic error at the city's edge is about 10 meters, while the systematic error at the city center is negligible. Thus, the motion trajectory analysis in this invention is based on a fast Fourier transform performed on a two-dimensional plane.

[0098] 5. Gaussian Blur

[0099] Gaussian blur is the convolution of an image with a Gaussian function. According to the convolution theorem, the Fourier transform of a Gaussian blur is the product of the Fourier transform of the image and the Fourier transform of the Gaussian function. Since the Fourier transform of the Gaussian function is another Gaussian function, a Gaussian blur acts as a low-pass filter for the image.

[0100] In noisy images, the noise is mainly distributed in the high-frequency Fourier modes. Therefore, Gaussian blur, as a low-pass filter, can reduce noise and improve the signal-to-noise ratio.

[0101] The two-dimensional centrally symmetric Gaussian kernel function is:

[0102]

[0103] Where w is the full width at half maximum (FWHM) of the Gaussian function. The image after Gaussian blurring is as follows:

[0104]

[0105] in, Let I(x,y) represent the convolution operation, and let I(x,y) represent the image. It is the image after Gaussian blurring, where G(x,y) is a two-dimensional centrally symmetric Gaussian kernel function.

[0106] With the widespread adoption of smart mobile devices and the development of wireless communication technology, massive amounts of user spatiotemporal trajectory data have been generated. Analyzing this data can reveal user movement patterns and further uncover population clustering patterns. For example, analyzing user spatiotemporal trajectory information can enrich user profiles, predict user behavior, and reveal potential user needs, playing a crucial role in commercial promotion such as precise personalized marketing and urban governance.

[0107] Currently, mainstream positioning technologies include GPS positioning, wireless fidelity (Wi-Fi) positioning, and mobile communication base station positioning. GPS positioning achieves location by receiving GPS satellite signals through a GPS chip built into the terminal, offering high accuracy. Wi-Fi positioning relies on the location of connected Wi-Fi devices; its accuracy is affected by the Wi-Fi signal coverage, which is typically on the order of hundreds of meters, resulting in lower positioning accuracy. Base station positioning is similar to Wi-Fi, determining location based on the terminal's connection to a base station. However, because base station coverage is much larger than Wi-Fi, the positioning accuracy of a single base station is lower than that of Wi-Fi.

[0108] Analyzing a user's movement trajectory essentially involves analyzing the location data from the user's mobile device to determine their movement path. While GPS positioning data is accurate, it also suffers from various limitations due to its radio-based nature. These limitations include interference from radio waves, tall buildings, and walls, leading to data gaps and incomplete data. Furthermore, GPS requires user authorization to activate and requires continuous searching for and recording of GPS satellite signals, resulting in significant power consumption. Users often disable GPS for battery life or privacy reasons, compromising the integrity of the data and hindering accurate trajectory analysis.

[0109] Mobile communication base station positioning technology is based on signaling data generated when a terminal interacts with a base station. As long as the mobile phone is powered on and using a mobile network signal, the signaling data is automatically uploaded to the telecom operator's server without user authorization or additional power consumption. Furthermore, mobile network signals largely cover residential areas, including inside buildings and underground parking garages, effectively mitigating data loss issues. However, base station signal strength is not constant, meaning the connection between the terminal and the base station is not fixed. When the base station signal strength changes drastically, the signal received by the same terminal at the same location at adjacent times may switch back and forth between multiple base stations—a phenomenon known as "ping-pong switching." In this case, it's impossible to determine which base station the terminal is closer to. Additionally, base station signals have a limited coverage area, and this coverage varies depending on the standard and frequency. For example, a 700MHz 5G base station can cover up to 10 kilometers, while a higher-frequency base station might only cover 250 meters. Therefore, when a terminal connects to a base station, it only knows that the terminal is within the base station's coverage area, not its exact location. In other words, signaling data alone cannot guarantee the accurate determination of the terminal's movement trajectory.

[0110] To address the aforementioned technical problems, this application provides a motion trajectory analysis method. This method acquires signaling data of a terminal within a target area, wherein the signaling data records the base stations connected to by the terminal. Based on the antenna patterns of all base stations within the target area, a base station model of the target area is established, which indicates the overall signal strength within the target area. Convolution-based motion trajectory analysis is performed based on the base station model and the terminal's signaling data to obtain the terminal's motion trajectory information. Since the base station model indicates the overall signal strength within the target area, by performing convolution-based motion trajectory analysis on the base station model and the terminal's signaling data, Gaussian blurring can be used to reduce noise in the signaling data and enhance its features, thereby accurately obtaining the motion trajectory.

[0111] Figure 2 A flowchart illustrating a motion trajectory analysis method provided in this application includes the following steps:

[0112] Step S110: Obtain signaling data of terminals within the target area.

[0113] The aforementioned signaling data is used to record the base stations connected to by the terminal. Signaling data is generated when the terminal interacts with a base station, recording which base station the terminal connected to at any given time while connected to the mobile network. As long as the terminal is powered on and connected to the mobile network, the signaling data is automatically uploaded to the server of the operator corresponding to the terminal; therefore, the signaling data can be directly obtained from the operator's server.

[0114] The target area refers to the region where users are managed. Specifically, since base station operations are typically at the city level, such as the city-level branch of a telecommunications operator, the target area can be a city-level region. When analyzing the movement trajectories of people to facilitate prevention and control, management, and other related work, the target area can be determined based on the specific needs of these efforts.

[0115] Step S120: Based on the antenna patterns of all base stations in the target area, establish a base station model for the target area.

[0116] The base station model described above is used to indicate the synthetic signal strength within the target area.

[0117] An antenna pattern is a graph showing how the relative field strength (normalized modulus) of the radiated field changes with direction at a certain distance from the antenna. By analyzing the antenna pattern of a base station, we can obtain the antenna radiation characteristics, such as the relationship between field strength amplitude, phase, polarization, and spatial angle. A complete antenna pattern is a three-dimensional spatial graph, plotted by measuring the radiation characteristics point by point on a sphere with a sufficiently large radius r, using the antenna phase center as the origin. Different types of patterns can be obtained based on the measured radiation characteristics; for example, measuring the field strength amplitude yields a field strength pattern, measuring the power yields a power pattern, measuring the polarization yields a polarization pattern, and measuring the phase yields a phase pattern.

[0118] As one implementation method, please refer to Figure 3 The step S120 of establishing a base station model for the target area based on the antenna patterns of all base stations within the target area includes the following steps (S121-S123):

[0119] Step S121: Obtain the antenna pattern of all base stations in the target area.

[0120] For example, there are N in the target area. s Each base station has a spatial power pattern of P. i (x,y,z), where i∈[1,N] s In practical applications, since the user's location can be considered to be on the ground, the base station power pattern used can be simplified to its projection on the ground, i.e., P. i (x,y,z)=P i (x,y,z=0).

[0121] Step S122: Perform Fourier transform on the antenna pattern of each base station to obtain the antenna pattern of each base station in Fourier space.

[0122] Based on the above base station example, we obtain N. s Antenna pattern P of each base station i (x,y), where i∈[1,N] s Therefore, performing a two-dimensional Fourier transform on the antenna pattern of each base station yields the following antenna pattern in Fourier space:

[0123]

[0124] Step S123: Linearly superimpose the antenna patterns of each base station in Fourier space to obtain the base station model described above.

[0125] For example, the base station model obtained after linear superposition is as follows:

[0126]

[0127] The base station model is obtained by linear superposition. This allows for direct modification of the established model even when base stations within the target area change, enabling faster and more convenient model updates. For existing base stations, as long as the signal transmitting and receiving units remain unchanged, including their orientation, the power patterns in both the actual 3D and Fourier spaces do not need to be altered and can be reused. In other words, the base station model obtained once can be reused if the base station remains unchanged. For new base stations, according to the convolution theorem, the Fourier transforms of the new base station's power pattern can be linearly superimposed onto the existing base station model. For closed base stations, the Fourier transforms of the closed base station's power pattern can be subtracted from the existing base station model.

[0128] Continue to refer to Figure 2 Step S130: Perform convolution-based motion trajectory analysis based on the base station model and the signaling data of the terminal to obtain the motion trajectory information of the terminal.

[0129] The base station model is used to indicate the overall signal strength within the target area. Therefore, by performing convolution-based motion trajectory analysis on the signaling data of the base station model and the terminal, Gaussian blurring can be used to reduce noise in the signaling data and enhance the features in the signaling data, thereby accurately obtaining the motion trajectory.

[0130] This application provides a motion trajectory analysis method, wherein the motion trajectory information includes the spatiotemporal trajectory of the aforementioned terminal; step S130, which involves performing convolution-based motion trajectory analysis based on the aforementioned base station model and the signaling data of the aforementioned terminal to obtain the motion trajectory information of the aforementioned terminal, can be specifically implemented through the following steps (S1311-S1315), please refer to... Figure 4 :

[0131] Step S1311: Determine the terminal location matrix of the above-mentioned terminal based on the above signaling data.

[0132] The terminal location matrix of the aforementioned terminal is a set of geographical locations of the base stations that the terminal has connected to. Since the geographical locations of the base stations are known, such as their latitude and longitude coordinates, the base stations that the terminal has connected to can be determined from the signaling data, and the terminal's location can be obtained based on the locations of the base stations it has connected to.

[0133] Step S1312: Convolve the above base station model as a convolution kernel to the above terminal position matrix to determine the approximate spatiotemporal trajectory of the above terminal.

[0134] Since the terminal's location is based on the locations of the base stations it has connected to, there is a significant discrepancy between the terminal's location matrix and its actual location. Furthermore, the signal strength between the terminal's location and the base stations depends on the relative positions of the terminal and the base stations. Therefore, by using the base station model, which indicates the overall signal strength within the target area, as a convolution kernel, the terminal's location determined based on the locations of the base stations it connects to can be corrected, thus obtaining an approximate spatiotemporal trajectory of the terminal.

[0135] Whether it's a base station handover that occurs at the same location via ping-pong switching, or a base station handover that occurs during the terminal's movement based on signal strength, these base station handover scenarios can represent the terminal's possible location information, thus enabling the terminal to be located based on the base station model.

[0136] For example, since a base station has a certain coverage area, when a terminal connects to that base station, it means the terminal is within the coverage area of ​​that base station, but its specific location is still unclear. The terminal switching base stations is largely based on the strength of the base station signal received by the terminal. Therefore, the base station model can be seen as the weights of the terminal at a certain location. By convolving the base station model, we can obtain the weighted values ​​of the terminal at various locations within these coverage areas, i.e., the approximate spatiotemporal trajectory L(x,y). According to the convolution theorem, L(x,y) can be written as:

[0137]

[0138] in, Representing the base station model, D(x,y) represents the terminal position matrix.

[0139] Step S1313: Determine the estimated spatiotemporal trajectory based on the above approximate spatiotemporal trajectory and the road position matrix within the above target area.

[0140] The aforementioned road location matrix is ​​the set of location coordinates of all roads within the target area. Since users typically move along roads while carrying their terminals, the road location matrix can be used to further refine the approximate spatiotemporal trajectory, thereby obtaining an estimated spatiotemporal trajectory that more accurately represents the user's actual movement trajectory.

[0141] It's worth noting that, assuming the roads remain unchanged, the road location matrix only needs to be constructed once and can be reused. The road location matrix can be generated through manual point marking, map reverse geocoding, or other methods. When actual roads change, such as road closures or the activation of newly built roads, linear additions and subtractions can be performed on the original road location matrix, making it more convenient and faster. Furthermore, the road location matrix only distinguishes between roads with and without roads, without differentiating specific road attributes, such as whether a road is a main road or a side road. Therefore, the road location matrix R(x,y) is a binary matrix, with elements representing roads being 1 and elements representing non-roads being 0.

[0142] L(x,y) represents the user's approximate spatiotemporal trajectory, i.e., a possible range of the actual spatiotemporal trajectory. This range may include unreasonable locations such as crossing water or airports. These unreasonable locations can be eliminated using a road location matrix. Specifically, the road location matrix is ​​a binary matrix that can be seen as a filter from the approximate spatiotemporal trajectory to the estimated spatiotemporal trajectory E(x,y), i.e., E(x,y) = R(x,y)·L(x,y).

[0143] Step S1314: Gaussian blur the estimated spatiotemporal trajectory and the road position matrix respectively to obtain the Gaussian blurred estimated spatiotemporal trajectory and the Gaussian blurred road position matrix.

[0144] Gaussian blur is the convolution of an image with a Gaussian function. According to the convolution theorem, the Fourier transform of a Gaussian blur is the product of the Fourier transform of the image and the Fourier transform of the Gaussian function. Since the Fourier transform of the Gaussian function is another Gaussian function, Gaussian blur acts as a low-pass filter for the image. Image noise is mainly distributed in the high-frequency Fourier modes; therefore, Gaussian blur, as a low-pass filter, reduces noise and improves the signal-to-noise ratio.

[0145] Specifically, the estimated spatiotemporal trajectory E(x,y) can be viewed as a set of location points oscillating around the true spatiotemporal trajectory M(x,y); that is, E(x,y) is M(x,y) with added noise. To achieve fast computation without loss of generality, the same Gaussian blur can be applied to all roads to reduce noise, where the full width at half maximum (FWHM) w of the Gaussian kernel function is taken as the average width of the road. The estimated spatiotemporal trajectory after Gaussian blurring is:

[0146]

[0147] The road position matrix after Gaussian blurring is:

[0148]

[0149] The two-dimensional centrally symmetric Gaussian kernel function is:

[0150]

[0151] Where w is the full width at half maximum (FWHM) of the Gaussian function.

[0152] Step S1315: Calculate the product of the estimated spatiotemporal trajectory after Gaussian blurring and the road position matrix after Gaussian blurring, and determine the spatiotemporal trajectory of the terminal based on the product.

[0153] The user's true spatiotemporal trajectory M(x,y) is the estimated spatiotemporal trajectory after Gaussian blurring. The set of center values ​​sliding along the road direction. Since the actual spacetime trajectory is unknown, it is unknown... Which path(s) should the user slide along? Therefore, the problem can be approximated as the user's actual spatiotemporal trajectory being the estimated spatiotemporal trajectory after Gaussian blurring. and product The set of maximum values ​​along the x-direction or y-direction, where

[0154] Therefore, by taking the maximum value along the x-direction or y-direction, the actual spatiotemporal trajectory M(x,y) of the terminal can be determined.

[0155] For example, if we take the maximum value along the x-direction, for a certain y value y j corresponding vector Its maximum value is at x i Position obtained, maximum value is Then we have:

[0156]

[0157] By iterating through all y values, the user's true spatiotemporal trajectory M(x,y) is obtained. Understandably, the maximum value can also be taken along the y-direction; the specific operation is the same as taking the maximum value along the x-direction, and will not be elaborated here.

[0158] As one implementation method, the signaling data also records the connection duration when the terminal connects to the base station; the motion trajectory information includes rest points; the step S130, which involves performing convolution-based motion trajectory analysis based on the base station model and the terminal's signaling data to obtain the terminal's motion trajectory information, can be specifically implemented through the following steps (S1321-S1323), please refer to... Figure 5 :

[0159] Step S1321: Determine the terminal location matrix of the terminal and the connection duration matrix between the terminal and the base station based on the above signaling data.

[0160] The terminal location matrix of the aforementioned terminal is a set of geographical locations of the base stations connected to the aforementioned terminal, and the connection duration matrix is ​​a set of connection durations between the aforementioned terminal and the base stations it has connected to. The terminal location matrix can be found in the specific implementation described above, and will not be repeated here. Since the signaling data records the time of connection to the base station, the difference between the time of connection of the next signaling data to the base station and the time of connection of the current signaling data to the base station can be considered as the connection duration of the terminal connecting to the base station corresponding to the current signaling data. Therefore, the connection duration matrix T(x,y) of the terminal connecting to the base station can be determined based on the signaling data, and the elements of T(x,y) that are 0 are consistent with D(x,y), that is, when the connection duration to the base station is 0, it indicates that the terminal is not connected to that base station, and the terminal's location is also 0 at this time.

[0161] Step S1322: Convolve the above base station model as a convolution kernel to the above connection duration matrix and the above terminal location matrix to determine the time-weighted approximate spatiotemporal trajectory of the above terminal based on the connection duration.

[0162] The aforementioned time-weighted approximate spatiotemporal trajectory is the set of connection durations between the terminal and the base station at various locations.

[0163] Using the duration T(x,y) of the terminal's connection to the base station as the weight of the terminal's location D(x,y), we obtain the time-weighted approximate spatiotemporal trajectory LT(x,y) based on the connection duration, specifically:

[0164]

[0165] Step S1323: Determine the stopping point of the terminal based on the predetermined stopping time threshold and the time-weighted approximate spatiotemporal trajectory.

[0166] LT(x,y) describes a dual-weighted localization method with base station power pattern as spatial weight and terminal connection duration as temporal weight. A dwell time threshold ε is predetermined. t If LT(x,y) is greater than the threshold, it indicates that the corresponding position (x,y) is a stopping point, that is:

[0167]

[0168] In practical applications, the location of the stationary point S(x,y) may be an empty area, which is contrary to common sense and may lead to errors. Therefore, in one possible implementation of this application, when determining the stationary point of the terminal based on the predetermined stationary time threshold and the time-weighted approximate spatiotemporal trajectory, it can be first determined whether the connection duration corresponding to the first position point in the time-weighted approximate spatiotemporal trajectory is greater than the stationary time threshold. If it is greater, the first position point is determined as a candidate stationary point. If the coordinates of the first position point A are (x1,y1), and the corresponding S(x1,y1) can be determined as a candidate stationary point according to the above method, then, points of interest within a preset stationary distance threshold are searched with the candidate stationary point as the search center. The points of interest are locations with pre-marked geographical attributes. If no points of interest exist, the candidate stationary point is determined as the stationary point of the terminal; if points of interest exist, the points of interest whose distance from the candidate stationary point is less than or equal to the preset distance threshold are determined as the stationary points of the terminal.

[0169] In conjunction with the first aspect above, in one possible implementation, when the aforementioned stopping point is a work / residence location, the aforementioned connection duration matrix is ​​a set of connection durations within a preset time range from the connection durations between the terminal and the base station determined based on the aforementioned signaling data.

[0170] Work-residence points, also known as workplaces and residences, are specific points of contact, typically within a certain time frame. Generally, people's residences are located between 0:00 and 6:00, while their workplaces are located between 10:00 and 16:00. This time frame is a hyperparameter, set according to specific circumstances. For example, for someone who works long-term night shifts, their workplace might be located between 0:00 and 6:00.

[0171] When determining a user's work-residence location, this can be achieved by modifying the duration of the terminal's connection to the base station. Specifically, the time range t is set. s To t e The modified duration matrix is ​​as follows:

[0172]

[0173] Similar to the steps for determining the stopping point described above, replace T(x,y) with T′(x,y) and set the threshold ε for the work-residence point. h This yields the corresponding LT′(x,y) and S′(x,y):

[0174]

[0175]

[0176] Here, H(x,y) represents the user's work and residence location.

[0177] In one possible implementation of this application, gathering points can also be identified, which play an important role in both business promotion and urban governance. In business promotion, knowing where people frequently gather, and the size or number of people at these gatherings, allows for targeted business operations and estimates of the required staffing levels. In urban governance, additional security personnel can be deployed to areas where people congregate; even in maintaining social order, if a normally quiet spot suddenly becomes a gathering point, it can be considered whether a mass incident has occurred, and timely security measures can be implemented.

[0178] Therefore, in the motion trajectory analysis method provided in this application, the aforementioned terminal is included in multiple terminals, and the method further includes the following process:

[0179] The first step is to determine the cluster location matrix corresponding to the above-mentioned multiple terminals and the cluster connection duration matrix between the above-mentioned multiple terminals and the base station based on the signaling data of the above-mentioned multiple terminals.

[0180] The cluster location matrix corresponding to the above-mentioned multiple terminals is a linear superposition of the terminal location matrix of each terminal among the multiple terminals. The terminal location matrix of the above-mentioned terminals is a set of geographical locations of the base stations connected to by the above-mentioned terminals. The cluster connection duration matrix is ​​a linear superposition of the normalized connection duration matrix of each terminal among the multiple terminals connected to the base station.

[0181] Crowd aggregation points are the intersections of the spatiotemporal trajectories of a large number of users. Identifying these intersections can be viewed as identifying the resident points of the crowd within their spatiotemporal trajectories. The cluster location matrix D of the crowd terminals... q (x,y) is a linear superposition of the terminal location matrices of each user:

[0182]

[0183] Where N p It represents the total number of users in the region.

[0184] For crowd gathering points, the focus is more on the number of people gathered there, rather than how long an individual stays at that location. Therefore, the cluster connection duration matrix T, involving multiple terminals connected to the base station within the crowd, is crucial. q (x, y) cannot be a linear superposition of the duration matrices of each user terminal's connection to the base station; otherwise, the contribution of one person staying at that location for 20 hours would be equivalent to the contribution of 20 people staying at that location for 1 hour. If the duration of stay is not considered, then waypoints might be treated as stopping points. Therefore, the cluster connection duration matrix T of multiple terminals connecting to the base station should be... q(x,y) is defined as the linear superposition of the normalized duration matrices of each user terminal's connection to the base station:

[0185]

[0186] Where max[T i [x,y] represents taking T. i The maximum value in (x,y).

[0187] The second step involves using the base station model as a convolution kernel to perform convolution on the cluster connection duration matrix and the cluster location matrix of the multiple terminals connected to the base station, in order to determine the approximate spatiotemporal trajectory of the multiple terminals.

[0188] The aforementioned cluster approximate spatiotemporal trajectory is a set of normalized connection durations of the multiple terminals at various locations connecting with the base station.

[0189] The cluster connection duration matrix T connecting multiple terminals to the base station q Using (x,y) as the weights of the terminal positions Dq(x,y), we obtain the cluster approximate spatiotemporal trajectory LTq(x,y) of multiple terminals:

[0190]

[0191] The third step is to determine whether the second position point in the above-mentioned cluster approximate spatiotemporal trajectory is a cluster point based on the predetermined clustering threshold and the above-mentioned cluster approximate spatiotemporal trajectory. The second position point is located on the above-mentioned cluster approximate spatiotemporal trajectory.

[0192] Specifically, it is determined whether the normalized connection duration corresponding to the second position point in the above cluster weighted approximate spatiotemporal trajectory is greater than the above clustering threshold. If it is greater, the above second position point is determined to be a clustering point.

[0193] For example, setting an aggregation threshold ε q If LT q If (x, y) is greater than the aggregation threshold, then the corresponding second position point (x, y) is the aggregation point, i.e.:

[0194]

[0195] As can be seen, the above mainly describes the technical solutions provided by the embodiments of this disclosure from a methodological perspective. To achieve the above functions, it includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the modules and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this disclosure can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware 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 disclosure.

[0196] This disclosure embodiment can divide the motion trajectory analysis device into functional modules according to the above method example. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this disclosure embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0197] like Figure 6 As shown, a motion trajectory analysis device 600 is provided in an embodiment of this disclosure. The device includes:

[0198] The signaling data acquisition module 610 is used to acquire signaling data of terminals within the target area; wherein the signaling data is used to record the base stations that the terminals have connected to.

[0199] The base station model building module 620 is used to build a base station model of the target area based on the antenna patterns of all base stations in the target area; the base station model is used to indicate the overall signal strength in the target area.

[0200] The motion trajectory information analysis module 630 is used to perform convolution-based motion trajectory analysis based on the base station model and the signaling data of the terminal to obtain the motion trajectory information of the terminal.

[0201] Optionally, the above base station model building module includes:

[0202] Antenna pattern acquisition unit, used to acquire antenna patterns of all base stations in the target area mentioned above;

[0203] The Fourier transform unit is used to perform Fourier transform on the antenna pattern of each base station to obtain the antenna pattern of each base station in Fourier space.

[0204] The base station model building unit is used to linearly superimpose the antenna patterns of each base station in Fourier space to obtain the base station model mentioned above.

[0205] Optionally, the aforementioned motion trajectory information includes the spatiotemporal trajectory of the aforementioned terminal; the aforementioned motion trajectory information analysis module includes:

[0206] The terminal location matrix determination unit is used to determine the terminal location matrix of the terminal based on the aforementioned signaling data; the terminal location matrix of the terminal is a set of geographical locations of the base stations that the terminal has connected to.

[0207] The approximate spatiotemporal trajectory determination unit is used to convolve the above-mentioned base station model as a convolution kernel onto the above-mentioned terminal position matrix to determine the approximate spatiotemporal trajectory of the above-mentioned terminal.

[0208] The estimated spatiotemporal trajectory determination unit is used to determine the estimated spatiotemporal trajectory based on the aforementioned approximate spatiotemporal trajectory and the road position matrix within the aforementioned target area; wherein the aforementioned road position matrix is ​​a set of position coordinates of all roads within the aforementioned target area;

[0209] Gaussian blur unit is used to perform Gaussian blur on the estimated spatiotemporal trajectory and the road position matrix respectively, so as to obtain the Gaussian blurred estimated spatiotemporal trajectory and the Gaussian blurred road position matrix.

[0210] The spatiotemporal trajectory determination unit is used to calculate the product of the estimated spatiotemporal trajectory after Gaussian blurring and the road position matrix after Gaussian blurring, and to determine the spatiotemporal trajectory of the terminal based on the product.

[0211] Optionally, the aforementioned signaling data also records the connection duration when the aforementioned terminal connects to the base station; the aforementioned motion trajectory information includes stopping points;

[0212] The aforementioned motion trajectory information analysis module includes:

[0213] The terminal location and duration determination unit is used to determine the terminal location matrix of the terminal and the connection duration matrix between the terminal and the base station based on the signaling data; wherein, the terminal location matrix of the terminal is a set of geographical locations of the base stations that the terminal has connected to, and the connection duration matrix is ​​a set of connection durations between the terminal and the base stations that it has connected to.

[0214] The time-weighted approximate spatiotemporal trajectory determination unit is used to convolve the above-mentioned base station model as a convolution kernel to the above-mentioned connection duration matrix and the above-mentioned terminal location matrix to determine the time-weighted approximate spatiotemporal trajectory of the above-mentioned terminal based on the connection duration; the above-mentioned time-weighted approximate spatiotemporal trajectory is the set of connection durations of the above-mentioned terminal connecting with the base station at each location;

[0215] The stopping point determination unit is used to determine the stopping point of the terminal based on a pre-determined stopping time threshold and the aforementioned time-weighted approximate spatiotemporal trajectory.

[0216] Optionally, the aforementioned stopping point determination unit includes:

[0217] The candidate stopping point determination subunit is used to determine whether the connection duration corresponding to the first position point in the above time-weighted approximate spatiotemporal trajectory is greater than the above stopping time threshold. If it is greater, the above first position point is determined as a candidate stopping point.

[0218] The point of interest search subunit is used to search for points of interest within a preset stopping distance threshold range with the above candidate stopping points as the search center. The above points of interest are locations that are pre-marked with geographical attributes.

[0219] The stopping point determination subunit is used to determine the candidate stopping point as the stopping point of the terminal if no point of interest exists; if a point of interest exists, the point of interest whose distance from the candidate stopping point is less than or equal to a preset distance threshold is determined as the stopping point of the terminal.

[0220] Optionally, if the aforementioned stopping point is a work and residence location, the aforementioned connection duration matrix is ​​a set of connection durations within a preset time range among the connection durations between the terminal and the base station determined based on the aforementioned signaling data.

[0221] Optionally, the terminal is included in multiple terminals, and the device further includes:

[0222] The cluster location and duration determination module is used to determine the cluster location matrix corresponding to the above-mentioned multiple terminals and the cluster connection duration matrix between the above-mentioned multiple terminals and the base station based on the signaling data of the above-mentioned multiple terminals; wherein the cluster location matrix corresponding to the above-mentioned multiple terminals is a linear superposition of the terminal location matrix of each terminal among the multiple terminals, the terminal location matrix of the above-mentioned terminals is a set of geographical locations of the base stations connected to by the above-mentioned terminals, and the cluster connection duration matrix is ​​a linear superposition of the normalized connection duration matrices between each terminal among the multiple terminals and the base station;

[0223] The cluster approximate spatiotemporal trajectory determination module is used to convolve the above base station model as a convolution kernel to the cluster connection duration matrix and the cluster location matrix of the above multiple terminals to determine the cluster approximate spatiotemporal trajectory of the above multiple terminals; the above cluster approximate spatiotemporal trajectory is the set of normalized connection durations of the above multiple terminals connecting with the base station at each location.

[0224] The aggregation point determination module is used to determine whether a second position point in the aforementioned cluster approximate spatiotemporal trajectory is an aggregation point based on a pre-determined aggregation threshold and the aforementioned cluster approximate spatiotemporal trajectory, wherein the aforementioned second position point is located on the aforementioned cluster approximate spatiotemporal trajectory.

[0225] Optionally, the above-mentioned aggregation point determination module includes:

[0226] The clustering judgment unit is used to determine whether the normalized connection duration corresponding to the second position point in the above cluster weighted approximate spatiotemporal trajectory is greater than the above clustering threshold. If it is greater, the second position point is determined to be a clustering point.

[0227] Through the above description of the embodiments, those skilled in the art will clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the device, module, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0228] This disclosure also provides a computer-readable storage medium storing instructions that, when executed by a computer, perform each step of the method flow shown in the above method embodiments.

[0229] Embodiments of this disclosure provide a computer program product containing instructions that, when executed on a computer, cause the computer to perform the motion trajectory analysis method described in the above method embodiments.

[0230] Embodiments of this disclosure provide a chip including a processor and a communication interface, the communication interface and the processor being coupled together, the processor being used to run computer programs or instructions to implement the motion trajectory analysis method as described in the above method embodiments.

[0231] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), registers, hard disks, optical fibers, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing, or any other form of computer-readable storage medium in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium may also be a component of the processor. The processor and the storage medium may reside in an application-specific integrated circuit (ASIC). In this embodiment of the disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0232] Since the apparatus, devices, computer-readable storage media, and computer program products in the embodiments of this disclosure can be applied to the above methods, the technical effects they can achieve can also be referred to the above method embodiments. The embodiments of this disclosure will not be repeated here.

[0233] In the embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units through some interfaces, and may be electrical, mechanical, or other forms.

[0234] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0235] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0236] The above description is merely a specific embodiment of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any changes or substitutions within the technical scope disclosed in this disclosure should be included within the scope of protection of this disclosure. Therefore, the scope of protection of this disclosure should be determined by the scope of the claims.

Claims

1. A method of motion trajectory analysis, characterized by, The method comprises: acquiring signaling data of a terminal in a target area; wherein the signaling data is used to record base stations connected by the terminal; establishing a base station model of the target area based on antenna patterns of all base stations in the target area; the base station model is used to indicate integrated signal strength in the target area; convolving the base station model as a convolution kernel with the signaling data of the terminal to obtain motion trajectory information of the terminal.

2. The method of claim 1, wherein, The step of establishing a base station model of the target area based on antenna patterns of all base stations in the target area comprises: acquiring antenna patterns of all base stations in the target area; performing Fourier transform on the antenna pattern of each base station respectively to obtain the antenna pattern of each base station in Fourier space; linearly superimposing the antenna pattern of each base station in Fourier space to obtain the base station model.

3. The method according to claim 1 or 2, characterized in that, The motion trajectory information comprises a space-time trajectory of the terminal; the step of convolving the base station model as a convolution kernel with the signaling data of the terminal to obtain motion trajectory information of the terminal comprises: determining a terminal position matrix of the terminal based on the signaling data; the terminal position matrix of the terminal is a set of geographical positions of base stations connected by the terminal; convolving the base station model as a convolution kernel with the terminal position matrix to determine an approximate space-time trajectory of the terminal; determining an estimated space-time trajectory based on the approximate space-time trajectory and a road position matrix in the target area; wherein the road position matrix is a set of position coordinates of all roads in the target area; performing Gaussian blurring on the estimated space-time trajectory and the road position matrix respectively to obtain a Gaussian-blurred estimated space-time trajectory and a Gaussian-blurred road position matrix; calculating the product of the Gaussian-blurred estimated space-time trajectory and the Gaussian-blurred road position matrix, and determining the space-time trajectory of the terminal based on the product.

4. The method according to claim 1 or 2, characterized in that, The signaling data also records the connection duration when the terminal connects a base station; the motion trajectory information comprises a stop point; The step of convolving the base station model as a convolution kernel with the signaling data of the terminal to obtain motion trajectory information of the terminal comprises: determining a terminal position matrix of the terminal and a connection duration matrix of the terminal and base stations based on the signaling data; wherein the terminal position matrix of the terminal is a set of geographical positions of base stations connected by the terminal, and the connection duration matrix is a set of connection durations between the terminal and base stations connected by the terminal; convolving the base station model as a convolution kernel with the connection duration matrix and the terminal position matrix to determine a time-weighted approximate space-time trajectory of the terminal based on connection duration; the time-weighted approximate space-time trajectory is a set of connection durations of the terminal at each position and base stations; determining a stop point of the terminal according to a predetermined stop time threshold and the time-weighted approximate space-time trajectory.

5. The method of claim 4, wherein, The step of determining a stop point of the terminal according to a predetermined stop time threshold and the time-weighted approximate space-time trajectory comprises: determining whether a connection duration corresponding to a first position point in the time-weighted approximate spatio-temporal trajectory is greater than the stopover time threshold, and if so, determining that the first position point is a candidate stopover point; searching for a point of interest within a preset stopover distance threshold range from the candidate stopover point, the point of interest being a position pre-marked with a geographic attribute; if there is no point of interest, determining that the candidate stopover point is a stopover point of the terminal; and if there is a point of interest, determining a point of interest with a distance from the candidate stopover point less than or equal to a preset distance threshold as the stopover point of the terminal.

6. The method of claim 4, wherein, In the case that the stopover point is a job or residence point, the connection duration matrix is a set of connection durations of the terminal and the base station within a preset time range, which is determined based on the signaling data.

7. The method according to claim 1 or 2, characterized in that, The terminal is included in a plurality of terminals, and the method further comprises: determining a cluster position matrix corresponding to the plurality of terminals and a cluster connection duration matrix of the plurality of terminals and the base station based on signaling data of the plurality of terminals; wherein the cluster position matrix corresponding to the plurality of terminals is a linear superposition of terminal position matrices of each terminal in the plurality of terminals, the terminal position matrix of the terminal is a set of geographic positions of the base stations connected by the terminal, and the cluster connection duration matrix is a linear superposition of normalized connection duration matrices of each terminal in the plurality of terminals and the base station; convolving the base station model as a convolution kernel with the cluster connection duration matrix of the plurality of terminals and the base station and the cluster position matrix to determine a cluster approximate spatio-temporal trajectory of the plurality of terminals; the cluster approximate spatio-temporal trajectory is a set of normalized connection durations of the plurality of terminals and the base station at each position; determining whether a second position point in the cluster approximate spatio-temporal trajectory is an aggregation point according to a predetermined aggregation threshold and the cluster approximate spatio-temporal trajectory, the second position being on the cluster approximate spatio-temporal trajectory.

8. A motion trajectory analysis apparatus characterized by comprising: The device comprises: a signaling data acquisition module configured to acquire signaling data of a terminal in a target area; wherein the signaling data is used to record base stations connected by the terminal; a base station model establishment module configured to establish a base station model of the target area based on an antenna pattern of all base stations in the target area; the base station model is used to indicate the integrated signal strength in the target area; a motion trajectory information analysis module configured to convolve the base station model as a convolution kernel with the signaling data of the terminal to obtain motion trajectory information of the terminal.

9. A motion trajectory analysis device characterized by comprising: comprise: a processor and a memory; wherein the memory is used to store computer execution instructions, and when the motion trajectory analysis device is running, the processor executes the computer execution instructions stored in the memory to enable the motion trajectory analysis device to perform the motion trajectory analysis method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores instructions, and when the instructions in the computer readable storage medium are executed by the processor of the motion trajectory analysis device, the motion trajectory analysis device performs the motion trajectory analysis method of any one of claims 1-7.

Images