Task switching method and device, equipment, computer storage medium and program product

Abstract

The invention discloses a task switching method and device, equipment, a computer storage medium and a program product. The method comprises the following steps: acquiring a task base station set needing to be processed in a task distribution period; under the condition that the task base station set contains the base stations which are not covered by the second probability matrix, taking the first probability matrix as a target probability matrix; under the condition that all the base stations in the task base station set are covered by the second probability matrix, taking the second probability matrix as a target probability matrix; generating a first base station task sequence based on the target probability matrix and the responsible area of the staff; and distributing a base station task corresponding to the first base station task sequence to the worker. According to the embodiment of the invention, the scientificity and accuracy of task distribution can be improved, and the orderliness and efficiency of task execution of the base station are ensured.

Description

Technical Field

[0001] This application belongs to the field of task switching technology, and in particular relates to a task switching method, apparatus, device, computer storage medium and program product. Background Technology

[0002] With the continuous evolution and large-scale deployment of mobile communication networks, especially the intensive construction of 5G networks, the number of wireless base stations that telecom operators need to manage and maintain has exploded. Against this backdrop, the scheduling efficiency and rationality of engineering construction field teams, responsible for base station construction, inspection, and fault handling, directly affect network construction progress, operation and maintenance costs, and service quality. How to achieve scientific, efficient, and balanced task scheduling between the massive, dynamic, and geographically dispersed base station tasks and limited field human resources has become a crucial and highly challenging technical issue in the field of network operations.

[0003] Currently, task scheduling for staff in the industry mainly relies on human experience or simple information tools. The common practice is for dispatchers to manually assign and coordinate tasks based on relatively static and limited information such as personal experience, base station location, and task urgency, via telephone or a simple task dispatch application. While this method achieves some degree of online task dispatch, its core scheduling logic still heavily depends on human judgment, representing a semi-automatic or even manual management model.

[0004] However, the aforementioned existing technical solutions have revealed several inherent technical problems in practical applications. First, manual scheduling struggles to quickly and accurately handle large-scale, multi-constraint, complex scheduling problems, leading to inefficient scheduling decisions and difficulty in ensuring the overall optimality of the solution. Second, due to a lack of systematic analysis and learning capabilities regarding historical task execution results (such as actual time consumption, path trajectory, and work score saturation), scheduling solutions often become disconnected from dynamically changing on-site conditions, failing to achieve continuous optimization based on real feedback. Furthermore, static scheduling rules cannot adapt to changes in network structure (such as the addition of new base stations) or fluctuations in personnel status, lacking flexibility. Ultimately, these problems easily lead to uneven work scores among staff, redundant round-trip journeys, and untimely task responses, hindering the improvement of overall network operation efficiency and quality. Summary of the Invention

[0005] This application provides a task switching method, apparatus, device, computer storage medium, and program product that can improve the scientificity and accuracy of task distribution and ensure the orderly and efficient execution of base station tasks.

[0006] On one hand, embodiments of this application provide a task switching method, the method comprising: obtaining a set of task base stations to be processed within a task distribution period; if the task base station set includes base stations not covered by a second probability matrix, using a first probability matrix as a target probability matrix; if all base stations in the task base station set are covered by the second probability matrix, using the second probability matrix as the target probability matrix; generating a first base station task sequence based on the target probability matrix and the area of ​​responsibility of the staff; and distributing base station tasks corresponding to the first base station task sequence to the staff; wherein, the first probability matrix is ​​a base station pair connection probability matrix generated based on base station location information and base station subset sampling data, wherein the sampling count value of each base station in the base station subset is greater than or equal to a preset threshold, and the sampling count value represents the total number of times the base station appears in all base station subsets; each The base station subset contains one central base station and multiple neighboring base stations. The central base station in each base station subset is selected from the set of base stations whose sample count value reaches the global minimum value at that sampling time. The multiple neighboring base stations in each base station subset are the multiple base stations that are closest to the central base station of the subset. The connection probability value in the first probability matrix is ​​calculated based on the frequency of each base station pair appearing as adjacent nodes in multiple optimal access paths determined by the base station subset sampling data. The second probability matrix is ​​obtained by adjusting the frequency of the corresponding base station pairs in the first probability matrix according to the difference between the second base station task sequence and the first base station task sequence. The second base station task sequence is obtained by randomly perturbing the first base station task sequence corresponding to multiple staff members based on the execution feedback data. The execution feedback data is obtained by multiple staff members executing their corresponding base station tasks.

[0007] In some possible implementations, before obtaining the set of task base stations to be processed within the task distribution period, the method further includes: obtaining execution feedback data of multiple staff members performing their corresponding base station tasks; based on the execution feedback data, randomly perturbing the first base station task sequence corresponding to multiple staff members to obtain a second base station task sequence; the random perturbation includes at least one of the following: swapping the staff members to which two base stations belong, transferring one base station from one staff member's sequence to another staff member's sequence, swapping the order of two base stations in one staff member's sequence; adjusting the frequency of corresponding base station pairs in the first probability matrix according to the difference between the second base station task sequence and the first base station task sequence to obtain a second probability matrix.

[0008] In some possible implementations, the execution feedback data includes the travel distance and work score corresponding to the first base station task sequence; based on the execution feedback data, the first base station task sequences corresponding to multiple staff members are randomly perturbed to obtain the second base station task sequence, including: randomly perturbing the first base station task sequence a preset number of times to obtain multiple third base station task sequences; for each third base station task sequence, the following steps are performed: calculating the total travel distance required to execute the third base station task sequence based on the travel distance of all staff members; determining the travel distance score corresponding to the total travel distance according to a first preset conversion rule; calculating the total working time required to execute the third base station task sequence based on the work scores of all staff members; determining the work score corresponding to the total working time according to a second preset conversion rule; weighted summing of the travel distance score and the work score to obtain the comprehensive score of the third base station task sequence; and selecting a candidate allocation scheme whose comprehensive score meets preset conditions from the multiple third base station task sequences as the second base station task sequence.

[0009] In some possible implementations, before obtaining the set of task base stations to be processed within the task distribution cycle, the method further includes: initializing a first-order matrix and a second-order matrix, where the elements in the i-th row and j-th column of the first-order matrix and the second-order matrix correspond to the base station pairs formed by the i-th base station and the j-th base station in a preset arrangement, and the initial value of each element is zero; for each different pair of base stations in all base stations, counting the total number of times the base station pair appears together in the same base station subset, and recording the total number as the element value of the corresponding position in the first-order matrix; counting the number of times each base station pair appears as adjacent nodes in the optimal access path, and recording the number of times adjacent nodes appear as the element value of the corresponding position in the second-order matrix; and constructing a first probability matrix based on the ratio of each element value in the second-order matrix to the corresponding element value in the first-order matrix.

[0010] In some possible implementations, based on the difference between the second base station task sequence and the first base station task sequence, the frequencies of corresponding base station pairs in the first probability matrix are adjusted to obtain a second probability matrix. This includes: obtaining newly added base station pair connections in the second base station task sequence, including base station pairs that are adjacent in the second base station task sequence but not adjacent in the first base station task sequence; obtaining removed base station pair connections in the first base station task sequence, including base station pairs that are adjacent in the first base station task sequence but not adjacent in the second base station task sequence; for each newly added base station pair connection, increasing the record value of its corresponding element in the first probability matrix to obtain an adjusted second probability matrix; for each removed base station pair connection, decreasing the record value of its corresponding element in the first probability matrix to obtain an adjusted second probability matrix; and recalculating the connection probability of each base station pair based on the adjusted second probability matrix and the first probability matrix to obtain the second probability matrix.

[0011] In some possible implementations, a first base station task sequence is generated based on the target probability matrix and the staff member's responsible area. This includes: for each staff member, determining the intersection of their responsible area and the task base station set, which serves as the staff member's set of base stations to be assigned; randomly selecting a base station from the set of base stations to be assigned as the starting base station of the task base station sequence; removing the starting base station from the set of base stations to be assigned to obtain the remaining base station set; using the starting base station as the current base station; repeating the following steps until the set of base stations to be assigned is empty, resulting in the first base station task sequence: calculating the distance from the current base station to each base station in the set of base stations to be assigned; obtaining the connection probability value corresponding to the current base station and each base station in the set of base stations to be assigned from the target probability matrix; multiplying the distance to the current base station and the corresponding connection probability value for each base station in the set of base stations to be assigned to obtain multiple comprehensive selection evaluation values; selecting the base station with the smallest comprehensive selection evaluation value from the set of base stations to be assigned as the target base station; removing the target base station from the set of base stations to be assigned and adding it to the end of the staff member's first base station task sequence; and updating the target base station to the current base station.

[0012] In some possible implementations, before obtaining the set of base stations to be processed within the task distribution cycle, the method further includes: for each base station subset, constructing a two-dimensional coordinate matrix containing the coordinates of all base stations in the subset based on the latitude and longitude coordinates of all base stations in the subset; determining the coordinate transformation matrix of the subset by extracting the extreme values ​​of the two-dimensional coordinate matrix in the longitude and latitude dimensions; performing a normalization transformation on the two-dimensional coordinate matrix based on the coordinate transformation matrix to obtain a normalized coordinate matrix; and using the coordinates in the normalized coordinate matrix as the base station location information of the subset.

[0013] On the other hand, embodiments of this application provide a task switching device, comprising: a set acquisition module for acquiring a set of task base stations to be processed within a task distribution period; a first selection module for selecting a first probability matrix as a target probability matrix when the task base station set includes base stations not covered by a second probability matrix; a second selection module for selecting a second probability matrix as a target probability matrix when all base stations in the task base station set are covered by the second probability matrix; a sequence generation module for generating a first base station task sequence based on the target probability matrix and the area of ​​responsibility of the staff; and a task distribution module for distributing base station tasks corresponding to the first base station task sequence to the staff; wherein the first probability matrix is ​​a base station pair connection probability matrix generated based on base station location information and base station subset sampling data, wherein the sampling count value of each base station in the base station subset is greater than or equal to a preset threshold, and the sampling count... The numerical representation represents the total number of times a base station appears in all base station subsets; each base station subset contains one central base station and multiple neighboring base stations; the central base station in each base station subset is selected from the set of base stations whose sampling count value reaches the global minimum value at that sampling time; the multiple neighboring base stations in each base station subset are the multiple base stations that are closest to the central base station of the subset; the connection probability value in the first probability matrix is ​​calculated based on the frequency of each base station pair appearing as adjacent nodes in multiple optimal access paths determined by the sampling data of the base station subset; the second probability matrix is ​​obtained by adjusting the frequency of corresponding base station pairs in the first probability matrix according to the difference between the second base station task sequence and the first base station task sequence; the second base station task sequence is obtained by randomly perturbing the first base station task sequence corresponding to multiple staff members based on the execution feedback data; the execution feedback data is obtained by multiple staff members executing their corresponding base station tasks.

[0014] In another aspect, embodiments of this application provide an electronic device, the device including: a processor and a memory storing computer program instructions; and a task switching method implemented by the processor when executing the computer program instructions.

[0015] In another aspect, embodiments of this application provide a computer storage medium on which computer program instructions are stored, and when the computer program instructions are executed by a processor, a task switching method is implemented.

[0016] In another aspect, embodiments of this application provide a computer program product in which instructions are executed by the processor of an electronic device, causing the electronic device to perform a task switching method.

[0017] The task switching method, apparatus, device, and computer storage medium of this application embodiment utilize a first probability matrix generated based on base station location information and sampled base station subset data. The base station subset sampling process ensures the sufficiency and representativeness of each base station's sampling, and its connection probability reflects the objective rationality of the base station as an adjacent node, providing a reliable foundation for task sequence generation. The second probability matrix incorporates optimization information from actual execution feedback data. This matrix is ​​selected for the task set of covered base stations, making task allocation more aligned with the actual execution scenario and improving task execution efficiency. Finally, by combining task allocation with the areas where staff are responsible, efficiency losses caused by cross-regional operations are avoided, achieving regionalization and rationalization of task allocation. This comprehensively improves the scientificity and accuracy of task distribution, ensuring the orderly and efficient execution of base station tasks. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a task switching method provided in one embodiment of this application; Figure 2 This is a flowchart illustrating a task switching method provided in another embodiment of this application; Figure 3 This is a flowchart illustrating a task switching method provided in yet another embodiment of this application; Figure 4 This is a flowchart illustrating a task switching method provided in yet another embodiment of this application; Figure 5 This is a schematic diagram of the structure of a task switching device provided in another embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in another embodiment of this application. Detailed Implementation

[0020] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0021] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0022] It should be noted that the acquisition, storage, use, and processing of data in this application embodiment all comply with the relevant provisions of national laws and regulations.

[0023] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0024] Existing scheduling models fail to meet the complex matching needs of "massive dynamic tasks and limited human resources" in communication wireless engineering construction scenarios, and lack systematic modeling and adaptive optimization capabilities for multi-dimensional constraints. On the one hand, with the large-scale deployment of communication networks, base station tasks exhibit characteristics of "massive quantity, geographical dispersion, diverse types, and dynamic additions." Manual scheduling or simple information tools can only rely on limited static information (such as geographical location and task urgency) for decision-making, making it impossible to quickly handle the combined optimization problem between "multiple tasks and multiple personnel," let alone balance multiple objectives such as shortest path, balanced workload, and regional responsibility matching. This results in scheduling schemes failing to achieve global optimization, leading to problems such as uneven workload and path redundancy. On the other hand, existing solutions lack effective utilization and learning mechanisms for historical execution data (such as actual workload, path trajectory, and time differences). They cannot correct scheduling deviations through data feedback, nor can they adapt to dynamic scenarios such as changes in network structure (such as the addition of new base stations) and fluctuations in personnel status. This keeps scheduling rules in a static and fixed state, unable to match with the actual situation on site in real time, ultimately resulting in low scheduling efficiency, delayed response, and failure to meet the efficiency and quality control requirements of network operations.

[0025] The task switching method in this application embodiment generates a first probability matrix based on base station location information and sampled base station subset data. The base station subset sampling process ensures the sufficiency and representativeness of each base station's sampling, and its connection probability reflects the objective rationality of the base station as an adjacent node, providing a reliable foundation for task sequence generation. The second probability matrix incorporates optimization information from actual execution feedback data. This matrix is ​​selected for the task set of covered base stations, making task allocation more aligned with the actual execution scenario and improving task execution efficiency. Finally, by combining task allocation with the areas where staff are responsible, efficiency losses caused by cross-regional operations are avoided, achieving regionalization and rationalization of task allocation. This comprehensively improves the scientificity and accuracy of task distribution, ensuring the orderly and efficient execution of base station tasks.

[0026] To address the problems of the prior art, embodiments of this application provide a task distribution method, apparatus, device, computer storage medium, and computer program product. The task distribution method provided in this application embodiment will be described first below.

[0027] Figure 1 A flowchart illustrating a task distribution method according to an embodiment of this application is shown. Figure 1 As shown, the method includes the following steps: S101, Obtain the set of base stations for tasks that need to be processed within the task distribution period.

[0028] As an example, the task distribution cycle can refer to a preset fixed time unit used to define the time range for task allocation. It can be set according to the actual operational needs of communication wireless engineering construction (such as daily or weekly) to ensure the regularity and timeliness of task allocation.

[0029] As an example, the task base station set can refer to the set of all base stations that need to be constructed, inspected, or troubleshooted during the current task distribution cycle. The set contains basic data such as the location information and task type of each base station.

[0030] Specifically, as an example, a communication network operation and management platform can be used to aggregate all pending base station task information within the current task distribution cycle, collect key information such as the location data of each base station and the construction and maintenance content to be performed, integrate and classify these base stations to form a structured task base station set, and provide complete data input for subsequent probability matrix selection and task sequence generation.

[0031] Before introducing S102, let's first introduce the first probability matrix and the second probability matrix: The first probability matrix is ​​a base station pair connection probability matrix generated based on base station location information and base station subset sampling data. The sampling count value of each base station in the base station subset is greater than or equal to a preset threshold, and the sampling count value represents the total number of times the base station appears in all base station subsets. Each base station subset contains one central base station and multiple neighboring base stations. The central base station in each base station subset is selected from the set of base stations whose sampling count value reaches the global minimum value at that sampling time. The multiple neighboring base stations in each base station subset are the multiple base stations that are closest to the central base station of the subset. The connection probability value in the first probability matrix is ​​calculated based on the frequency of each base station pair appearing as adjacent nodes in multiple optimal access paths determined by the base station subset sampling data.

[0032] The second probability matrix is ​​obtained by adjusting the frequencies of corresponding base station pairs in the first probability matrix based on the difference between the second base station task sequence and the first base station task sequence. The second base station task sequence is obtained by randomly perturbing the first base station task sequence corresponding to multiple staff members based on the execution feedback data. The execution feedback data is obtained by multiple staff members executing their corresponding base station tasks.

[0033] As an example, the first probability matrix can be constructed based on the location information of all base stations in the network and under construction, and the sampling data of a subset of base stations. It is used to reflect the probability that any two base stations forming a pair of base stations will appear as adjacent nodes in the optimal access path, with only the geographical distance between the base stations as the core consideration factor.

[0034] As an example, the second probability matrix can be based on the first probability matrix. According to the difference between the second base station task sequence and the first base station task sequence, the matrix formed after the occurrence frequency of the corresponding base station pair is adjusted, and historical task execution feedback data is incorporated to take into account both path rationality and adaptability to actual construction scenarios.

[0035] As an example, base station subset sampling data can be obtained by sampling the full number of base stations multiple times. Each subset contains a central base station and multiple neighboring base stations, and the sampling count value of each base station in all subsets is not lower than a preset threshold.

[0036] As an example, the sample count can be used to count the total number of times a single base station appears in all base station subsets, ensuring that the sampling process covers all base stations and avoiding data deviation due to insufficient sampling of some base stations.

[0037] As an example, a central base station can be selected from a set of base stations whose sample count value reaches the global minimum value at the time of sampling. Its geographical location has the advantage of radiating to the surrounding area, which facilitates the establishment of efficient connections with surrounding base stations.

[0038] As an example, neighboring base stations can be multiple base stations that are closest to the central base station, used to construct a local network topology for a subset of base stations, providing basic data for calculating the optimal access path.

[0039] As an example, the optimal access path can be obtained by solving the open path traveling salesman problem using a dynamic programming algorithm based on the location information of a subset of base stations. This can achieve the goal of traversing all base stations in the subset from the starting base station with the shortest travel distance.

[0040] As an example, the second base station task sequence can be formed by randomly perturbing the first base station task sequences corresponding to multiple staff members based on execution feedback data. It is an optimized adjustment result of the initial allocation scheme.

[0041] As an example, execution feedback data can be the actual data generated by staff after completing base station tasks, including the distance traveled, working time, and quality of task completion, providing real and effective data support for optimizing allocation schemes.

[0042] S102, if the task base station set includes base stations not covered by the second probability matrix, the first probability matrix is ​​used as the target probability matrix.

[0043] Specifically, as an example, each base station in the task base station set can be verified to determine whether it exists in the base station list corresponding to a row or column of the second probability matrix, i.e., to confirm whether the base station is covered by the second probability matrix. If the verification finds that there is at least one base station in the task base station set that is not covered by the second probability matrix, it means that the base station lacks corresponding historical execution feedback data, and the second probability matrix cannot provide it with an accurate connection probability reference. In this case, the first probability matrix is ​​determined as the target probability matrix, relying on the base station location information to ensure the basic rationality of task allocation.

[0044] S103, if all base stations in the task base station set are covered by the second probability matrix, the second probability matrix is ​​used as the target probability matrix.

[0045] Specifically, as an example, a comprehensive verification can be performed on all base stations in the task base station set to confirm that each base station exists in the base station list corresponding to a row or column of the second probability matrix, meaning that all base stations are covered by the second probability matrix. This indicates that these base stations have accumulated sufficient historical execution feedback data, and the second probability matrix can provide a connection probability reference that is more in line with the actual construction scenario. Therefore, the second probability matrix is ​​determined as the target probability matrix, providing a more accurate basis for the subsequent task sequence generation.

[0046] S104, Based on the target probability matrix and the area of ​​responsibility of the staff, generate the first base station task sequence.

[0047] As an example, the area of ​​responsibility for staff can be pre-defined as a fixed geographical work area for each field worker, which clarifies the boundaries of staff responsibilities and ensures that task allocation is compatible with area management.

[0048] As an example, the first base station task sequence can be generated by combining the connection probability of the target probability matrix with the area of ​​responsibility of each staff member, based on an ordered list of base station tasks for each staff member. This can achieve the dual objectives of optimal travel distance and matching of area responsibilities.

[0049] As one implementation of S104, S104 may also include the following steps: For each staff member, determine the intersection of their responsible area and the set of task base stations, which will be used as the set of base stations to be assigned to that staff member.

[0050] As an example, the set of base stations to be assigned can refer to the overlapping part of the staff member's area of ​​responsibility and the set of base stations for tasks in the current task distribution period, that is, the range of base station tasks that the staff member can undertake in the current period that conform to the division of responsibilities within the area.

[0051] Specifically, as an example, the geographical information of each staff member's responsible area can be extracted, including precisely defined data such as latitude and longitude ranges. Simultaneously, the latitude and longitude coordinates of each base station in the task base station set are obtained. Through geospatial intersection operations, all base stations that are both within the staff member's responsible area and belong to the task base station set are selected. These selected base stations are then integrated to form the staff member's pending base station set, ensuring that task allocation aligns with regional management requirements and avoiding travel redundancy caused by cross-regional dispatching.

[0052] Randomly select a base station from the set of base stations to be assigned as the starting base station of the task base station sequence.

[0053] As an example, the starting base station can refer to the first base station in the task base station sequence to execute a task. It is the starting point for constructing the task execution order, and its selection directly affects the planning efficiency of subsequent task paths.

[0054] Specifically, as an example, a random sampling algorithm can be used to select a base station as the starting base station from all base station data in the set of base stations to be assigned. Random selection avoids the problem of simplistic path planning caused by a fixed starting point. Furthermore, combined with optimization in subsequent steps, it ensures the overall path's rationality and adapts to task execution requirements in different base station distribution scenarios.

[0055] Remove the starting base station from the set of base stations to be allocated to obtain the remaining set of base stations.

[0056] As an example, the remaining base station set can refer to the set of all base stations remaining after removing the starting base station from the base station set to be allocated, which is the range of base stations to be gradually included in the task sequence.

[0057] Specifically, as an example, after determining the starting base station, this base station is removed from the data source of the set of base stations to be assigned through data processing methods, and the remaining base stations automatically form the remaining base station set. This operation ensures that the starting base station is not repeatedly included in the task sequence, guarantees the uniqueness and orderliness of task allocation, and lays the data foundation for subsequent iterations to generate task sequences.

[0058] Use the starting base station as the current base station.

[0059] As an example, the current base station can refer to the base station that has been determined to be included in the sequence and serves as the reference benchmark for subsequent base station selection during the iterative process of task sequence generation. It is the core node for calculating the subsequent path connection relationship.

[0060] Specifically, as an example, after selecting the starting base station and determining the set of remaining base stations, the selected starting base station is marked as the current base station. This marking operation can be achieved through data indexing, enabling subsequent steps to quickly retrieve data such as the location information of this base station and its connection probability with other base stations, providing a clear reference benchmark for path optimization calculations.

[0061] Repeat the following steps until the set of base stations to be assigned is empty, to obtain the first base station task sequence: Calculate the distance from the current base station to each base station in the set of base stations to be assigned.

[0062] As an example, base station distance can refer to the relative distance between the current base station and each base station in the set of base stations to be assigned. It is calculated based on the normalized coordinates of the base stations and is used to reflect the geographical proximity between base stations.

[0063] Specifically, as an example, normalized coordinate data of each base station in the current base station and the set of base stations to be assigned can be retrieved. This coordinate data is obtained by extracting the extreme values ​​of the base station's latitude and longitude to construct a coordinate transformation matrix, and then performing a normalization transformation. Based on the normalized coordinates, the distance between the current base station and each base station to be assigned is calculated using the Euclidean distance calculation formula, obtaining the distance value of each base station to be assigned relative to the current base station. This provides geospatial data support for the subsequent calculation of the comprehensive selection evaluation value.

[0064] Obtain the connection probability value corresponding to the current base station and each base station in the set of base stations to be assigned from the target probability matrix.

[0065] As an example, the connection probability value can refer to the probability that a base station pair consisting of the current base station and a base station to be assigned, recorded in the target probability matrix, appears as adjacent nodes in the optimal access path, reflecting the possibility that the base station pair is preferentially selected as a continuous operation path.

[0066] Specifically, as an example, the target probability matrix serves as a structured storage medium for the connection probabilities of base stations, with each row and column corresponding to a different base station. Based on the identification information of the current base station and the base station to be assigned, the intersection of the corresponding row and column in the target probability matrix is ​​located. The value corresponding to this intersection is the connection probability value between the current base station and the base station to be assigned. If the target probability matrix is ​​a first probability matrix, this value is calculated based on a subset of base station sampling data and the optimal access path frequency; if it is a second probability matrix, this value has been adjusted and optimized by incorporating historical execution feedback data.

[0067] For each base station in the set of base stations to be assigned, multiply it by the distance to the current base station and the corresponding connection probability value to obtain multiple comprehensive selection evaluation values.

[0068] As an example, the comprehensive selection evaluation value can refer to a quantitative evaluation index that integrates the geographical distance between base stations and the connection probability. It is used to measure the rationality of using a certain base station to be assigned as the next working node of the current base station. The smaller the value, the more suitable the base station is as the next base station to perform the task.

[0069] Specifically, as an example, for each base station in the set of base stations to be assigned, the calculated distance value is multiplied by the obtained connection probability value to obtain the comprehensive selection evaluation value corresponding to that base station. This calculation method takes into account both geographical proximity and the historical rationality (or basic rationality) of the path selection, enabling the evaluation results to comprehensively reflect the core needs of path optimization and avoid path redundancy caused by single-dimensional decision-making.

[0070] From the set of base stations to be allocated, select the base station with the smallest comprehensive selection evaluation value as the target base station.

[0071] As an example, the target base station can refer to the base station that is most suitable as the next working node of the current base station, selected based on the comprehensive selection evaluation value in the current iteration process, and is the core node to be included in the task sequence in the future.

[0072] Specifically, as an example, the comprehensive selection evaluation values ​​corresponding to all base stations to be assigned are sorted, and the base station corresponding to the lowest evaluation value is selected as the target base station. Selecting the base station with the lowest comprehensive selection evaluation value ensures that the next operation path is shorter in geographical distance and more reasonable in path selection, gradually building the overall optimal task execution sequence.

[0073] Remove the target base station from the set of base stations to be assigned and add it to the end of the worker's first base station task sequence.

[0074] As an example, data processing operations remove the target base station from the set of base stations to be allocated, updating the data source of the set. Simultaneously, the target base station's identification information, location information, and other relevant data are added to the end of the first base station task sequence, completing the iterative expansion of the task sequence. This operation ensures both the accuracy of the set of base stations to be allocated and the orderly construction of the task sequence, guaranteeing that each base station is allocated only once and the execution order is clear.

[0075] Update the target base station to the current base station.

[0076] As an example, after removing the target base station and adding the task sequence, the target base station is remarked as the current base station, providing a new reference benchmark for the next round of iteration. By continuously updating the current base station, the task sequence is gradually extended until the set of base stations to be assigned is empty, ultimately forming a complete first base station task sequence.

[0077] The task distribution method in this application determines the set of base stations to be assigned based on the area where the staff is responsible, thus achieving regional division of tasks, reducing cross-regional operation costs, and improving the convenience of task execution. Secondly, by randomly selecting the starting base station, the flexibility of the task sequence is increased, avoiding the local optima problem caused by a fixed starting point. Furthermore, by calculating the evaluation value by combining the distance from the current base station to each candidate base station with the corresponding connection probability, the economic efficiency of path distance and the rationality of connection relationships are considered, effectively reducing invalid travel. Finally, by iteratively selecting the base station with the smallest comprehensive evaluation value to construct the sequence, the generated task sequence is ensured to achieve optimal path length and execution efficiency, significantly improving the staff's work efficiency and reducing task execution costs.

[0078] S105 distributes the base station tasks corresponding to the first base station task sequence to the staff.

[0079] As an example, task distribution, which transmits the generated first base station task sequence to the staff terminal, is a key link connecting task planning and actual execution.

[0080] Specifically, as an example, a task dispatch system can push the first base station task sequence corresponding to each staff member to their dedicated terminal. After receiving the task information, the terminal displays key information such as the execution order of the base station tasks, the latitude and longitude coordinates of each base station, and construction and maintenance requirements in a visual format, allowing staff members to execute tasks in an orderly manner according to the sequence, ensuring the effective implementation of the task allocation plan.

[0081] The task distribution method in this application embodiment generates a first probability matrix based on base station location information and sampled base station subset data. The base station subset sampling process ensures the sufficiency and representativeness of each base station's sampling, and its connection probability reflects the objective rationality of base stations as adjacent nodes, providing a reliable foundation for task sequence generation. The second probability matrix incorporates optimization information from actual execution feedback data. This matrix is ​​selected for the task set of covered base stations, making task allocation more aligned with the actual execution scenario and improving task execution efficiency. Finally, by combining task allocation with the areas where staff are responsible, efficiency losses caused by cross-regional operations are avoided, achieving regionalization and rationalization of task allocation. This comprehensively improves the scientificity and accuracy of task distribution, ensuring the orderly and efficient execution of base station tasks.

[0082] As another implementation of this application, in order to ensure that the connection probabilities in the matrix consistently match the actual execution effect, such as Figure 2 As shown, before S101, the method may also include the following steps.

[0083] S201, Obtain execution feedback data from multiple staff members performing their corresponding base station tasks; Specifically, as an example, data related to each worker's base station tasks is automatically collected through data interaction between the worker's terminal and the communication network operation and management platform. Travel distance data, based on the terminal's positioning function, records the actual distance a worker travels from one base station to another during task execution, and is accurately calculated using the base station's normalized coordinates. Work duration data records the worker's arrival time at the base station and task completion time through the terminal, calculating the execution time of a single base station task and the total duration of the entire task sequence. Task completion quality data is generated after the worker submits their task completion status on the terminal, and is verified and confirmed by the backend system. This data is categorized and summarized according to dimensions such as worker, task sequence, and base station to form a structured execution feedback dataset, ensuring data integrity and usability.

[0084] S202, based on the execution feedback data, the task sequences of the first base stations corresponding to multiple staff members are randomly perturbed to obtain the task sequences of the second base stations. The random perturbation includes at least one of the following: swapping the staff members to which the two base stations belong, transferring one base station from one staff member's sequence to another staff member's sequence, or swapping the order of the two base stations within one staff member's sequence.

[0085] As an example, random perturbation can refer to the random adjustment operation performed on the generated first base station task sequence in order to explore a better task allocation scheme. By changing the staff assigned to the base station or the execution order, multiple differentiated task sequences are generated, providing a basis for subsequent selection of the optimal scheme.

[0086] As one implementation of S202, the execution feedback data includes the travel distance and work score corresponding to the first base station task sequence. S202 may also include the following steps: The task sequence of the first base station is randomly perturbed a preset number of times to obtain multiple task sequences of the third base station.

[0087] As an example, the preset number of perturbation operations can be pre-set based on the task scale, base station distribution density, and optimization requirements of the wireless communication engineering construction, ensuring that a sufficient number of differentiated task sequences are generated to select the optimal solution.

[0088] As an example, the third base station task sequence can refer to the candidate task sequence formed after a single random perturbation of the first base station task sequence. It serves as an intermediate carrier for exploring better allocation schemes, and each sequence has differentiated adjustments from the original first base station task sequence.

[0089] Specifically, as an example, a reasonable preset number of iterations can be set based on the size of the current task base station set, the number of staff, and historical optimization data. This ensures that sufficient optimization possibilities are covered while avoiding excessive computational resource waste caused by excessive perturbations. Subsequently, the first base station task sequence is adjusted multiple times using a preset random perturbation method. The perturbation methods include swapping the staff belonging to two base stations, transferring one base station from one staff member's sequence to another staff member's sequence, and swapping the order of two base stations within one staff member's sequence. Each perturbation uses only one method or a combination of multiple methods, and each perturbation generates an independent third base station task sequence. This perturbation operation is repeated until the preset number of iterations is reached, forming a set of third base station task sequences containing multiple differentiated candidate schemes.

[0090] For each third base station task sequence, perform the following steps: Calculate the total travel distance required to execute the third base station task sequence based on the travel distance of all staff performing their tasks.

[0091] As an example, travel distance can refer to the geographic distance that a worker travels from one base station to the next during the execution of a single base station task sequence. It is calculated based on the normalized coordinates of the base stations and reflects the path length of the task execution.

[0092] As an example, the total travel distance can refer to the sum of the travel distances traveled by all staff members to complete their assigned base station tasks when executing a certain third base station task sequence. It is a core indicator for measuring the degree of path optimization of the task sequence.

[0093] Specifically, as an example, for each third base station task sequence, the base station execution order corresponding to each worker can be extracted. Based on the normalized coordinate data of the base stations, the Euclidean distance calculation formula is used to calculate the travel distance between adjacent base stations in each worker's task sequence, and the summation is used to obtain the travel distance of that worker executing that sequence. The travel distances of all workers are summed to obtain the total travel distance corresponding to the third base station task sequence, ensuring that the calculation result can fully reflect the total path length of the entire task allocation scheme.

[0094] Based on the first preset conversion rule, determine the travel distance score corresponding to the total travel distance.

[0095] As an example, the first preset conversion rule can refer to a rule that is pre-defined based on the efficiency requirements and path optimization goals of wireless communication engineering construction, and converts the total travel distance into a quantitative score for intuitive measurement of the path optimization effect.

[0096] As an example, the trip distance score can refer to the quantitative score obtained by converting the total trip distance based on the first preset conversion rule. The higher the score, the better the path optimization of the third base station task sequence.

[0097] Specifically, as an example, the first preset conversion rule divides the total travel distance into different intervals and assigns corresponding scores. Specifically, the rule could be: a total travel distance less than or equal to 60 corresponds to 10 points; greater than 60 but less than or equal to 80 corresponds to 8 points; greater than 80 but less than or equal to 100 corresponds to 6 points; greater than 100 but less than or equal to 120 corresponds to 4 points; greater than 120 but less than or equal to 140 corresponds to 2 points; and greater than 140 corresponds to 0 points. For each third base station task sequence, the total travel distance is matched to its corresponding distance interval, and the corresponding score is extracted as the travel distance score for that sequence, thus achieving a quantitative evaluation of the total travel distance.

[0098] Calculate the total working time required to execute the third base station task sequence based on the work scores of all staff members performing the tasks.

[0099] As an example, work performance rating can refer to a quantitative indicator that reflects the efficiency of staff in performing a single base station task. It is determined based on data such as actual task execution time and task completion quality, and serves as the basis for deriving the total working time.

[0100] As an example, total working time can refer to the sum of working time required for all staff to complete their assigned base station tasks when executing a certain third base station task sequence. It is a key indicator for measuring the workload balance and execution efficiency of the task sequence.

[0101] Specifically, as an example, a correspondence between work scores and single-base station task execution time can be pre-established. Based on historical execution feedback data, the average task execution time corresponding to different work scores can be determined. For each third base station task sequence, all base stations in each worker's task sequence are extracted. Based on the work score corresponding to each base station, the estimated execution time of each base station is queried, and the work time of the worker executing that sequence is accumulated. The work times of all workers are summed to obtain the total work time corresponding to the third base station task sequence, ensuring that the time calculation closely matches the actual execution efficiency.

[0102] Based on the second preset conversion rule, determine the work score corresponding to the total working time.

[0103] As an example, the second preset conversion rule can refer to a rule that is pre-formulated based on the workload balance requirements and operational efficiency targets of communication wireless engineering construction, which converts the total working time into a quantitative score, and is used to intuitively measure the rationality of workload allocation.

[0104] Specifically, as an example, the second preset conversion rule uses the average working time of all staff members performing the third base station task sequence as a benchmark to divide intervals and assign corresponding scores. Specifically, the rule is as follows: a difference of less than or equal to 1 hour between a single staff member's working time and the average corresponds to 10 points; more than 1 hour but less than or equal to 2 hours corresponds to 8 points; more than 2 hours but less than or equal to 3 hours corresponds to 6 points; more than 3 hours but less than or equal to 4 hours corresponds to 4 points; more than 4 hours but less than or equal to 5 hours corresponds to 2 points; and more than 5 hours corresponds to 0 points. For each third base station task sequence, the average working time of all staff members is calculated, and then the interval to which the difference between each staff member's working time and the average belongs is matched. The average score of all staff members is taken as the work score for that sequence, achieving a quantitative evaluation of the balance between total working time and workload.

[0105] The comprehensive score of the third base station task sequence is obtained by weighted summation of the travel distance score and the work score.

[0106] As an example, the comprehensive score can refer to a quantitative evaluation index that integrates the travel distance score and the workload score. It comprehensively reflects the path optimization degree, workload balance and execution efficiency of the third base station task sequence through a weighted summation method, and is the core basis for selecting the optimal solution.

[0107] Specifically, as an example, the weight of the travel distance score can be set based on the importance of path optimization and workload balancing in wireless communication engineering construction. The weight of the travel distance score can be set at 60%, and the weight of the workload score at 40%. For each third base station task sequence, its travel distance score is multiplied by 60%, and its workload score is multiplied by 40%. The two are added together to obtain the comprehensive score of the sequence. The comprehensive score is as high as 20 points and as low as 0 points. The higher the score, the more the task sequence meets the actual operational needs.

[0108] From multiple third base station task sequences, the candidate allocation scheme that meets the preset conditions in terms of comprehensive score is selected as the second base station task sequence.

[0109] As an example, preset conditions can refer to comprehensive scoring and screening criteria pre-set according to the optimization objectives of communication wireless engineering construction, which are used to determine the optimal candidate task sequence.

[0110] Specifically, as an example, the preset condition is to select the candidate scheme with the highest comprehensive score. If there are multiple sequences with the same highest comprehensive score, the travel distance score or work score can be further considered, or one can be randomly selected. The comprehensive scores of all third base station task sequences are sorted, and candidate schemes that meet the preset conditions are extracted and determined as second base station task sequences, providing a basis for subsequent adjustment of the probability matrix and optimization of subsequent task allocation.

[0111] The task distribution method in this application specifies the execution feedback data as travel distance and work score, focusing on the core efficiency indicators of task execution and making the optimization direction more precise. Multiple third base station task sequences are generated through a preset number of random perturbations, ensuring the comprehensiveness of the optimization exploration and avoiding the omission of better solutions. Furthermore, the total travel distance and total work time are converted into quantitative scores through first and second preset conversion rules, and a comprehensive score is obtained by weighted summation, realizing a quantifiable evaluation of the task solutions and avoiding interference from subjective judgment. Based on the comprehensive score, solutions that meet the conditions are selected as the second base station task sequence, ensuring that the selected solutions achieve an optimal balance between travel distance and work efficiency, further improving the overall efficiency of task execution.

[0112] S203, based on the difference between the second base station task sequence and the first base station task sequence, adjust the frequency of the corresponding base station pair in the first probability matrix to obtain the second probability matrix.

[0113] As an example, a base station pair can refer to a combination of any two base stations. It is the object of connection probability calculation in the probability matrix, reflecting the association relationship between the two base stations as adjacent execution nodes.

[0114] As an example, frequency can refer to the number of times a base station pair appears as adjacent nodes in a task sequence. It is the core data for calculating the connection probability of a base station pair, and changes in frequency directly affect the magnitude of the connection probability.

[0115] Specifically, as an example, the second base station task sequence can be compared with the first base station task sequence to identify newly added base station pairs and removed base station pairs. Subsequently, the matrix recording the number of times adjacent base station pairs appear is adjusted, with the number of times corresponding to newly added connections increased by 1 and the number of times corresponding to removed connections decreased by 1. Then, the connection probability of each base station pair is recalculated by combining the matrix of the total number of times base station pairs appear together, and the second probability matrix can be obtained.

[0116] The task distribution method in this application collects feedback data from staff performing tasks, ensuring that the adjustment of the probability matrix is ​​supported by real data. This avoids deviations caused by purely theoretical calculations and makes subsequent task allocation more consistent with actual operating conditions. Optimizing the task sequence of the first base station using multiple random perturbation methods fully explores the possibilities of different task allocation schemes, overcoming the limitations of the initial sequence. Simultaneously, adjusting the frequency of the first probability matrix based on sequence differences enables dynamic iterative updates of the probability matrix, ensuring that the connection probabilities in the matrix continuously align with actual execution results. This provides a better data foundation for selecting the subsequent target probability matrix, gradually improving the optimization level of task distribution.

[0117] As another implementation of this application, in order to objectively reflect the reasonableness of the base station as an adjacent node, such as Figure 3 As shown, before S101, the method may also include the following steps.

[0118] S301, initialize the first number matrix and the second number matrix. The elements in the i-th row and j-th column of the first number matrix and the second number matrix correspond to the base station pairs formed by the i-th base station and the j-th base station in the preset arrangement order. The initial value of each element is zero.

[0119] As an example, the first number matrix can refer to a matrix used to record the total number of times each pair of different base stations co-occurs in the same base station subset. The rows and columns of the matrix are arranged in a preset order corresponding to each base station, and the values ​​of the matrix elements directly reflect the co-occurrence frequency of the corresponding base station pair.

[0120] As an example, the second-order matrix can refer to a matrix used to record the number of times each base station pair appears as an adjacent node in the optimal access path. The row and column arrangement rules of the matrix are the same as those of the first-order matrix, and the values ​​of the matrix elements directly reflect the frequency of the corresponding base station pair appearing as an adjacent node.

[0121] As an example, i and j are both index numbers of base stations, used to uniquely identify a single base station in a preset arrangement order. Specifically, i corresponds to the base station associated with the i-th row of the matrix, and j corresponds to the base station associated with the j-th column of the matrix. The element in the i-th row and j-th column of the matrix specifically corresponds to the base station pair formed by the i-th and j-th base stations in the preset arrangement order. By using the indexes i and j, relevant statistical data (such as co-occurrence count and adjacent occurrence count) of this base station pair can be accurately associated with and recorded.

[0122] As an example, the preset arrangement order can refer to the arrangement order of all base stations in the network and under construction in advance according to specific rules (such as base station number, geographical region) in order to standardize matrix construction and ensure the consistency of base station indexes.

[0123] As an example, the initial value can refer to the initial value set at the beginning of the construction of the frequency matrix before the statistical frequency of each element has been accumulated, in order to ensure the accuracy and standardization of subsequent statistical operations.

[0124] Specifically, as an example, we can organize all base station information, both currently in operation and under construction, and number and sort all base stations according to a preset order, clearly defining the row and column index positions of each base station in the matrix. The matrix size is determined based on the total number of base stations. If the total number of base stations is N, then both the first and second number matrices are N×N two-dimensional matrices. Subsequently, all elements of both matrices are initialized, with each element's initial value uniformly set to zero. This lays the data foundation for subsequent statistics on the co-occurrence and adjacent occurrence counts of base station pairs, ensuring that the statistical process accumulates from the initial state and avoiding interference from initial values ​​in the statistical results.

[0125] S302, for each pair of different base stations in all base stations, count the total number of times that the base station pair appears together in the same base station subset, and record the total number of times as the element value of the corresponding position in the first number matrix.

[0126] As an example, the total number of co-occurrences can refer to the cumulative number of times that two base stations in a base station pair are simultaneously included in the same base station subset, reflecting the co-occurrence frequency of the base station pair during the base station sampling process.

[0127] Specifically, as an example, we can iterate through all base station pairs consisting of two different base stations in the entire database. For each base station pair, we retrieve all generated base station subsets one by one. We determine whether both base stations of the base station pair exist simultaneously in the currently retrieved base station subset. If they do, we accumulate the co-occurrence count of the base station pair. After completing the retrieval of all base station subsets, we fill the corresponding row and column index positions of the base station pair in the first-order matrix. That is, the element value of the i-th row and j-th column is updated to the total co-occurrence count of the base station pair, ensuring that the matrix elements accurately reflect the co-occurrence frequency of each base station pair.

[0128] S303, count the number of times each base station appears as an adjacent node in the optimal access path, and record the number of times the adjacent node appears as the element value of the corresponding position in the second number matrix.

[0129] As an example, the optimal access path can refer to the path obtained by solving the open path traveling salesman problem using a dynamic programming algorithm based on the location information of each subset of base stations, which can achieve the goal of traversing all base stations in the subset from the starting base station with the shortest travel distance.

[0130] As an example, adjacent nodes can refer to two base stations that appear consecutively in the optimal access path, that is, two base stations that staff members visit in sequence according to the path when performing a task, corresponding to a continuous journey in the path.

[0131] As an example, the number of adjacent occurrences can refer to the cumulative number of times a base station pair appears as an adjacent node in the optimal access path of all base station subsets, reflecting the rationality of the base station pair as a continuous operation node.

[0132] Specifically, as an example, for the optimal access path corresponding to each subset of base stations, all consecutive adjacent base station pairs in the path can be extracted. That is, according to the path execution order, the k-th base station and the (k+1)-th base station are sequentially paired as adjacent base station pairs. The optimal access paths of all base station subsets are traversed, and the count of each extracted adjacent base station pair is accumulated. After traversing all paths, the cumulative number of adjacent occurrences of each base station pair is filled into the row and column index position of the base station pair in the second number matrix. That is, the value of the element in the i-th row and j-th column is updated to the number of adjacent occurrences of the base station pair, ensuring that the matrix elements accurately reflect the frequency of each base station pair as adjacent nodes.

[0133] S304. Construct the first probability matrix based on the ratio of each element value in the second probability matrix to the corresponding element value in the first probability matrix.

[0134] As an example, the ratio can refer to the quotient of the value of a certain element in the second number matrix and the value of the corresponding element in the first number matrix, directly representing the probability of the base station pair appearing as an adjacent node.

[0135] Specifically, as an example, we can iterate through all corresponding elements in the first and second probability matrices. For each element, if the value in the first probability matrix is ​​not zero, we calculate the ratio of the corresponding element value in the second probability matrix to that element value; if the value in the first probability matrix is ​​zero (i.e., the base station pair does not co-occur in any subset of base stations), we set the corresponding ratio to zero. We use the ratio calculated for each position as the element value of the corresponding position in the first probability matrix. After calculating all elements, we form the complete first probability matrix. Each element value in this matrix is ​​between 0 and 1. A larger value indicates a higher probability that the corresponding base station pair appears as adjacent nodes in the optimal access path, providing a probabilistic basis for path selection in subsequent task sequence generation.

[0136] The task distribution method in this application initializes two number matrices and standardizes the correspondence between matrix elements and base station pairs, providing a standardized data structure for subsequent statistics and ensuring the orderliness and accuracy of data records. The first number matrix counts the total number of co-occurrences of base station pairs, and the second number matrix counts the number of times base station pairs appear adjacently in the optimal access path. These two types of data reflect the association characteristics of base station pairs from different dimensions, providing a comprehensive basis for connection probability calculation. The first probability matrix is ​​constructed by the ratio of the two, making the connection probability calculation logic clear and the data support sufficient, ensuring that the matrix can objectively reflect the rationality of base station pairs as adjacent nodes, and providing a high-quality data foundation for the generation of the initial task sequence.

[0137] As one implementation of S203, S203 may also include the following steps: Acquire newly added base station pairs in the second base station task sequence. The newly added base station pairs include base station pairs that are adjacent in the second base station task sequence but not adjacent in the first base station task sequence.

[0138] As an example, a newly added base station pair connection can refer to a pair of base stations that are adjacent in the second base station task sequence, while the base station pairs that are not adjacent in the first base station task sequence are newly added combinations of continuously operating nodes after random perturbation optimization.

[0139] Specifically, as an example, all adjacent base station pairs are extracted from the first base station task sequence and the second base station task sequence, forming two sets of base station pairs. The set of base station pairs in the second base station task sequence is compared with the set of base station pairs in the first base station task sequence, and base station pairs that exist only in the second base station task sequence set are selected. These base station pairs are the newly added base station pair connections. The selection process is achieved by comparing the base station pair identifiers in the two sets one by one, ensuring that no newly added adjacent connections are missed.

[0140] Obtain the base station pairs that have been removed from the first base station task sequence. The removed base station pairs include those that are adjacent in the first base station task sequence but not adjacent in the second base station task sequence.

[0141] As an example, the removed base station pair connection can refer to a base station pair that is adjacent in the first base station task sequence but not adjacent in the second base station task sequence. It is a combination of consecutive working nodes that is canceled after random perturbation optimization.

[0142] Specifically, as an example, using the two base station pair sets extracted above, the base station pair set of the first base station task sequence is compared with the base station pair set of the second base station task sequence. Base station pairs that exist only in the first base station task sequence set are selected; these base station pairs are the removed base station pair connections. By accurately comparing the base station pair identifiers, all canceled adjacent connection relationships are fully identified.

[0143] For each newly added base station pair connection, the record value of its corresponding element in the first exponent matrix is ​​added to obtain the adjusted second exponent matrix.

[0144] As an example, the adjusted second-order matrix can refer to the matrix formed by adding or removing base stations to adjust the connection count of the original second-order matrix, and its element values ​​are more in line with the optimized task execution scenario.

[0145] As an example, the recorded value can refer to the statistical count of the corresponding base station pair in the frequency matrix, reflecting the co-occurrence frequency or adjacent occurrence frequency of the base station pair.

[0146] Specifically, as an example, for each newly added base station pair connection, the row index i and column index j of the corresponding base station pair in the first exponential matrix can be determined according to the preset arrangement order of the base stations. The element in the i-th row and j-th column of the first exponential matrix is ​​found, and its current record value is increased by 1, thus reflecting the actual situation that the base station pair is more suitable as an adjacent working node after optimization. After adjusting the record values ​​of all newly added base station pair connections, the initially adjusted second exponential matrix is ​​obtained.

[0147] For each removed base station pair connection, the record value of its corresponding element in the first exponent matrix is ​​reduced to obtain the adjusted second exponent matrix.

[0148] Specifically, as an example, for each removed base station pair connection, based on the preset arrangement order of the base stations, the row index i and column index j corresponding to that base station pair in the first frequency matrix are determined. The element in the i-th row and j-th column of the first frequency matrix is ​​found, and its current record value is reduced by 1, thus reflecting the actual situation that the base station pair is not suitable as an adjacent working node after optimization. After adjusting the record values ​​of all removed base station pair connections, the final adjusted second frequency matrix is ​​obtained, ensuring that the matrix element values ​​accurately reflect the frequency of adjacent occurrences of the optimized base station pairs.

[0149] Based on the adjusted second probability matrix and the first probability matrix, the connection probability of each base station pair is recalculated to obtain the second probability matrix.

[0150] Specifically, as an example, the process iterates through all corresponding elements in the adjusted second-order matrix and the first-order matrix. For each element, if the element value in the first-order matrix is ​​not zero, the ratio of the corresponding element value in the adjusted second-order matrix to that element value is calculated; if the element value in the first-order matrix is ​​zero, the corresponding ratio is set to zero. The ratio calculated for each position is used as the element value of the corresponding position in the second probability matrix. After calculating all elements, a complete second probability matrix is ​​formed. This matrix can provide a more accurate basis for selecting adjacent nodes in subsequent task allocation, improving the efficiency and rationality of task execution.

[0151] The task distribution method in this application accurately identifies newly added and removed base station connections, clarifies the objects to be adjusted in the probability matrix, and ensures that the adjustment direction does not deviate from the actual optimization requirements. For newly added connections, corresponding element record values ​​are added; for removed connections, corresponding record values ​​are reduced, so that the adjustment of the frequency matrix directly corresponds to the sequence optimization results, achieving precision in the adjustment operation. Based on the adjusted second frequency matrix, the connection probabilities are recalculated, enabling the second probability matrix to quickly absorb optimization experience from task execution, ensuring that the dynamic updates of the matrix are more targeted, and further improving the rationality and efficiency of subsequent task allocation schemes.

[0152] As another implementation of this application, in order to make the calculation of the relative distance between base stations more accurate, such as Figure 4 As shown, before S101, the method may also include the following steps.

[0153] S401. For each subset of base stations, construct a two-dimensional coordinate matrix containing the coordinates of all base stations in that subset, based on the latitude and longitude coordinates of all base stations in the subset.

[0154] As an example, latitude and longitude coordinates refer to the precise location data of a base station in geographic space. They consist of longitude and latitude and are the core information representing the geographical location of the base station.

[0155] As an example, a two-dimensional coordinate matrix can refer to a data structure that stores the latitude and longitude coordinates of all base stations in a single base station subset in matrix form. Each row of the matrix corresponds to one base station, and each column corresponds to longitude and latitude respectively, which facilitates subsequent coordinate transformation and calculation.

[0156] Specifically, as an example, for each generated subset of base stations, the raw latitude and longitude coordinates of all base stations in that subset are collected to ensure the accuracy and completeness of the coordinate data. The base stations in the subset are sorted according to a preset order, and the longitude value of each base station is used as the first column element of a matrix row, and the latitude value as the second column element, constructing a two-dimensional coordinate matrix with 2 columns and the number of rows equal to the size of the base station subset. For example, if a base station subset contains m base stations, the two-dimensional coordinate matrix is ​​an m×2 matrix, with each row uniquely corresponding to the latitude and longitude coordinates of one base station, providing a structured data foundation for subsequent coordinate transformations.

[0157] S402, by extracting the extreme values ​​of the two-dimensional coordinate matrix in the longitude and latitude dimensions, the coordinate transformation matrix of the base station subset is determined.

[0158] As an example, longitude extrema can refer to the maximum and minimum longitude values ​​of all base stations in a two-dimensional coordinate matrix, used to define the geographical range of the subset of base stations in the longitude direction.

[0159] As an example, the latitude dimension extrema can refer to the maximum and minimum latitude values ​​of all base stations in a two-dimensional coordinate matrix, used to define the geographical range of the subset of base stations in the latitudinal direction.

[0160] As an example, a coordinate transformation matrix can refer to a parameter matrix constructed based on the extreme values ​​of latitude and longitude dimensions for coordinate normalization transformation, which includes the scaling ratio and offset of longitude and latitude to ensure the uniformity and accuracy of coordinate transformation.

[0161] Specifically, as an example, the longitude column of the two-dimensional coordinate matrix can be traversed to extract the maximum and minimum longitude values, which are denoted as the maximum longitude value and the minimum longitude value, respectively. Similarly, the latitude column can be traversed to extract the maximum and minimum latitude values, which are denoted as the maximum latitude value and the minimum latitude value, respectively. Based on the extracted extreme values, coordinate transformation parameters are calculated: the longitude scaling factor is 1 divided by the difference between the maximum and minimum longitude values, and the latitude scaling factor is 1 divided by the difference between the maximum and minimum latitude values; the longitude offset is the minimum longitude value, and the latitude offset is the minimum latitude value. These parameters are integrated to form a coordinate transformation matrix, which contains the scaling factor and offset for both longitude and latitude, providing a calculation basis for subsequent normalization transformations.

[0162] S403, based on the coordinate transformation matrix, performs a normalization transformation on the two-dimensional coordinate matrix to obtain a normalized coordinate matrix.

[0163] As an example, normalization transformation can refer to the process of converting the latitude and longitude coordinates of a base station from the original geographic coordinate range to a uniform standard range (between 0 and 1), which is used to eliminate the impact of differences in the geographic range of different subsets of base stations on distance calculation.

[0164] As an example, a normalized coordinate matrix can refer to a base station coordinate matrix obtained after normalization transformation, in which all coordinate values ​​are between 0 and 1, facilitating path distance calculation and comparison across base station subsets.

[0165] Specifically, as an example, based on the coordinate transformation matrix determined in step S402, the coordinates of each base station in the two-dimensional coordinate matrix are normalized. For the longitude value of each base station, the formula "Normalized Longitude = (Original Longitude - Longitude Offset) × Longitude Scaling Ratio" is used for calculation; for the latitude value of each base station, the formula "Normalized Latitude = (Original Latitude - Latitude Offset) × Latitude Scaling Ratio" is used for calculation. The calculated normalized longitude and latitude are then used to replace the original longitude and latitude values ​​in the two-dimensional coordinate matrix to form a normalized coordinate matrix. For example, if the minimum longitude value of a subset of base stations is 1 and the maximum longitude value is 7, then the longitude scaling ratio is 1 / 6. If the original longitude of a base station is 2, its normalized longitude is (2-1) × 1 / 6 = 1 / 6, ensuring that all normalized coordinate values ​​are between 0 and 1, thus achieving coordinate standardization.

[0166] S404 uses the coordinates in the normalized coordinate matrix as the base station location information for this subset of base stations.

[0167] As an example, base station location information can refer to information that, after normalization, represents the relative position of a base station in a local subset of geographic space. Compared to the original latitude and longitude coordinates, it is more suitable for calculating the relative distance between base stations and for path planning.

[0168] Specifically, as an example, after normalization, the normalized longitude and latitude corresponding to each base station in the normalized coordinate matrix represent the relative position information of that base station within the current subset. This normalized coordinate matrix is ​​stored as the base station position information for that subset, and subsequent operations such as calculating the relative distance between base stations and solving for the optimal access path are all based on this normalized position information. Through normalization, the inconsistency in coordinate scale caused by geographical differences between different base station subsets is eliminated, ensuring the uniformity and accuracy of distance calculations between base stations within each subset, and providing reliable location data support for subsequent path planning and probability matrix construction.

[0169] The task distribution method in this application constructs a two-dimensional coordinate matrix based on the latitude and longitude coordinates of base stations, realizing structured storage of location information and providing a standardized data format for subsequent processing. Secondly, by extracting the extreme values ​​of the latitude and longitude dimensions to determine the coordinate transformation matrix, the relevance and accuracy of the normalization transformation are ensured, adapting to the geographical differences of different base station subsets. Finally, the two-dimensional coordinate matrix is ​​normalized to unify the coordinates to a standard range, eliminating distance calculation deviations caused by geographical scale differences among different base station subsets. This makes the calculation of relative distances between base stations more accurate, providing reliable location data support for solving the optimal access path and constructing the probability matrix, and ensuring the accuracy of subsequent task distribution-related calculations.

[0170] Based on the task distribution method provided in the above embodiments, this application also provides specific implementations of the task distribution device. Please refer to the following embodiments.

[0171] First see Figure 5 The task distribution device 50 provided in this application embodiment includes the following modules: The set acquisition module 501 is used to acquire the set of task base stations that need to be processed within the task distribution period; The first selection module 502 is used to select the first probability matrix as the target probability matrix when the task base station set includes base stations not covered by the second probability matrix. The second selection module 503 is used to select the second probability matrix as the target probability matrix when all base stations in the task base station set are covered by the second probability matrix. The sequence generation module 504 is used to generate the first base station task sequence based on the target probability matrix and the area of ​​responsibility of the staff. Task distribution module 505 is used to distribute base station tasks corresponding to the first base station task sequence to staff. The first probability matrix is ​​a base station pair connection probability matrix generated based on base station location information and base station subset sampling data. The sampling count value of each base station in the base station subset is greater than or equal to a preset threshold, and the sampling count value represents the total number of times the base station appears in all base station subsets. Each base station subset contains one central base station and multiple neighboring base stations. The central base station in each base station subset is selected from the set of base stations whose sampling count value reaches the global minimum value at that sampling time. The multiple neighboring base stations in each base station subset are the multiple base stations that are closest to the central base station of the subset. The connection probability value in the first probability matrix is ​​calculated based on the frequency of each base station pair appearing as adjacent nodes in multiple optimal access paths determined by the base station subset sampling data. The second probability matrix is ​​obtained by adjusting the frequency of the corresponding base station pairs in the first probability matrix based on the difference between the second base station task sequence and the first base station task sequence. The second base station task sequence is obtained by randomly perturbing the first base station task sequence corresponding to multiple staff members based on the execution feedback data. The execution feedback data is obtained by multiple staff members executing their corresponding base station tasks.

[0172] In some embodiments, the task switching device 50 may further include the following modules: The feedback acquisition module is used to acquire execution feedback data from multiple staff members performing their corresponding base station tasks. The perturbation generation module is used to randomly perturb the first base station task sequence corresponding to multiple staff members based on execution feedback data to obtain the second base station task sequence; the random perturbation includes at least one of the following: swapping the staff members to which the two base stations belong, transferring one base station from the sequence of one staff member to the sequence of another staff member, and swapping the order of two base stations in the sequence of one staff member.

[0173] In some embodiments, the execution feedback data includes the travel distance and work score corresponding to the first base station task sequence; the disturbance generation module includes: The multiple perturbation unit is used to randomly perturb the first base station task sequence a preset number of times to obtain multiple third base station task sequences. The scoring calculation unit performs the following operations for each third base station task sequence: The distance calculation subunit is used to calculate the total travel distance required to execute the third base station task sequence based on the travel distance of all staff performing tasks. The first scoring subunit is used to determine the travel distance score corresponding to the total travel distance according to the first preset conversion rule; The time calculation subunit is used to calculate the total working time required to execute the third base station task sequence based on the work scores of all staff members. The second scoring subunit is used to determine the work score corresponding to the total working time according to the second preset conversion rule; The comprehensive scoring subunit is used to perform a weighted summation of the travel distance score and the work score to obtain the comprehensive score of the third base station task sequence; The sequence selection subunit is used to select a candidate allocation scheme that meets the preset conditions in terms of comprehensive score from multiple third base station task sequences as the second base station task sequence.

[0174] The matrix adjustment module is used to adjust the frequency of the corresponding base station pair in the first probability matrix according to the difference between the second base station task sequence and the first base station task sequence, so as to obtain the second probability matrix.

[0175] In some embodiments, the disturbance generation module includes: The matrix initialization unit is used to initialize the first-order matrix and the second-order matrix. The elements in the i-th row and j-th column of the first-order matrix and the second-order matrix correspond to the base station pairs formed by the i-th base station and the j-th base station in a preset arrangement order. The initial value of each element is zero. The first statistical unit is used to count the total number of times that the base station pair appears together in the same base station subset for each different pair of base stations in all base stations, and record the total number of times as the element value of the corresponding position in the first statistical matrix; The second statistical unit is used to count the number of times each base station appears as an adjacent node in the optimal access path, and record the number of times the adjacent node appears as the element value of the corresponding position in the second statistical matrix. The probability calculation unit is used to construct the first probability matrix based on the ratio of each element value in the second number matrix to the corresponding element value in the first number matrix.

[0176] In some embodiments, the matrix adjustment module includes: A new connection unit is added to acquire newly added base station pairs in the second base station task sequence. The newly added base station pairs include base station pairs that are adjacent in the second base station task sequence but not adjacent in the first base station task sequence. The removal connection unit is used to obtain the base station pair connections that have been removed in the first base station task sequence. The removed base station pair connections include base station pairs that are adjacent in the first base station task sequence but not adjacent in the second base station task sequence. The exponent increment unit is used to add the record value of the corresponding element in the first exponent matrix for each newly added base station pair connection, so as to obtain the adjusted second exponent matrix; The number reduction unit is used to reduce the recorded value of the corresponding element in the first number matrix for each removed base station pair connection, so as to obtain the adjusted second number matrix; The second probability unit is used to recalculate the connection probability of each base station pair based on the adjusted second probability matrix and the first probability matrix, thus obtaining the second probability matrix.

[0177] In some embodiments, the sequence generation module includes: The assignment determination unit is used to determine the intersection of the area under the responsibility of each worker and the set of task base stations, which is used as the set of base stations to be assigned to that worker. The initial selection unit is used to randomly select a base station from the set of base stations to be allocated as the starting base station of the task base station sequence; The remaining acquisition unit is used to remove the starting base station from the set of base stations to be allocated, and obtain the set of remaining base stations; The current setting unit is used to set the starting base station as the current base station; The iterative generation unit is used to repeatedly perform the following operations until the set of base stations to be allocated is empty, thus obtaining the first base station task sequence: The distance calculation subunit is used to calculate the distance from the current base station to each base station in the set of base stations to be allocated; The probability acquisition subunit is used to obtain the connection probability value between the current base station and each base station in the set of base stations to be allocated from the target probability matrix; The evaluation value calculation subunit is used to multiply each base station in the set of base stations to be allocated by the distance to the current base station and the corresponding connection probability value to obtain multiple comprehensive selection evaluation values; The target selection subunit is used to select the base station with the smallest comprehensive selection evaluation value from the set of base stations to be allocated as the target base station; The sequence update subunit is used to remove the target base station from the set of base stations to be assigned and add it to the end of the worker's first base station task sequence; the current update subunit is used to update the target base station to the current base station.

[0178] In some embodiments, the task switching device 50 may further include the following modules: The coordinate matrix unit is used to construct a two-dimensional coordinate matrix containing the coordinates of all base stations in each base station subset, based on the latitude and longitude coordinates of all base stations in the subset. The transformation matrix module is used to determine the coordinate transformation matrix of the base station subset by extracting the extreme values ​​of the two-dimensional coordinate matrix in the longitude and latitude dimensions. The normalization module is used to perform a normalization transformation on a two-dimensional coordinate matrix based on the coordinate transformation matrix to obtain a normalized coordinate matrix; The location determination module is used to use the coordinates in the normalized coordinate matrix as the base station location information of the subset of base stations.

[0179] Figure 6 A schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application is shown.

[0180] The electronic device may include a processor 601 and a memory 602 storing computer program instructions.

[0181] Specifically, the processor 601 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0182] Memory 602 may include mass storage for data or instructions. For example, and not limitingly, memory 602 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 602 may include removable or non-removable (or fixed) media. Where appropriate, memory 602 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 602 is non-volatile solid-state memory.

[0183] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the methods according to one aspect of this disclosure.

[0184] The processor 601 implements any of the task switching methods described in the above embodiments by reading and executing computer program instructions stored in the memory 602.

[0185] In one example, the electronic device may also include a communication interface 603 and a bus 610. For example, Figure 6 As shown, the processor 601, memory 602, and communication interface 603 are connected through bus 610 and complete communication with each other.

[0186] The communication interface 603 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0187] Bus 610 includes hardware, software, or both, that couples components of an electronic device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 610 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0188] Furthermore, in conjunction with the task switching methods in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the task switching methods in the above embodiments.

[0189] This application also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the task switching methods described in the above embodiments.

[0190] It should be clarified that this application 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 this application 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 this application.

[0191] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they 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 application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0192] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0193] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0194] The above are merely specific embodiments of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A task distribution method, characterized in that, include: Obtain the set of task base stations that need to be processed within the task distribution period; If the set of task base stations includes base stations not covered by the second probability matrix, the first probability matrix shall be used as the target probability matrix. If all base stations in the task base station set are covered by the second probability matrix, the second probability matrix shall be used as the target probability matrix. Based on the target probability matrix and the area of ​​responsibility of the staff, a first base station task sequence is generated; Distribute the base station tasks corresponding to the first base station task sequence to the staff; The first probability matrix is ​​a base station pair connection probability matrix generated based on base station location information and base station subset sampling data. The sampling count value of each base station in the base station subset is greater than or equal to a preset threshold, and the sampling count value represents the total number of times the base station appears in all base station subsets. Each base station subset includes a central base station and multiple neighboring base stations. The central base station in each base station subset is selected from the set of base stations whose sampling count value reaches the global minimum value at that sampling time. The multiple neighboring base stations in each base station subset are the multiple base stations closest to the central base station of that subset. The connection probability value in the first probability matrix is ​​calculated based on the frequency with which each base station pair appears as an adjacent node in multiple optimal access paths determined by the base station subset sampling data. The second probability matrix is ​​obtained by adjusting the frequency of the corresponding base station pair in the first probability matrix according to the difference between the second base station task sequence and the first base station task sequence. The second base station task sequence is obtained by randomly perturbing the first base station task sequence corresponding to multiple staff members based on the execution feedback data. The execution feedback data is obtained by the multiple staff members executing their corresponding base station tasks.

2. The method according to claim 1, characterized in that, Before obtaining the set of task base stations to be processed within the task distribution period, the method further includes: Obtain execution feedback data from multiple staff members performing their corresponding base station tasks; Based on the execution feedback data, the first base station task sequence corresponding to the multiple staff members is randomly perturbed to obtain the second base station task sequence; the random perturbation includes at least one of the following: swapping the staff members to which two base stations belong, transferring one base station from one staff member's sequence to another staff member's sequence, or swapping the order of two base stations in one staff member's sequence; Based on the difference between the second base station task sequence and the first base station task sequence, the frequencies of the corresponding base station pairs in the first probability matrix are adjusted to obtain the second probability matrix.

3. The method according to claim 2, characterized in that, The execution feedback data includes the travel distance and work score corresponding to the first base station task sequence; the step of randomly perturbing the first base station task sequences corresponding to the multiple workers based on the execution feedback data to obtain the second base station task sequence includes: The first base station task sequence is randomly perturbed a preset number of times to obtain multiple third base station task sequences; For each of the third base station task sequences, the following steps are performed: Based on the travel distances of all staff members performing their tasks, calculate the total travel distance required to execute the third base station task sequence; According to the first preset conversion rule, the travel distance score corresponding to the total travel distance is determined; Calculate the total working time required to execute the third base station task sequence based on the work scores of all staff members performing the tasks. According to the second preset conversion rule, the work score corresponding to the total working time is determined; The comprehensive score of the third base station task sequence is obtained by weighted summation of the trip distance score and the work score. From the plurality of third base station task sequences, a candidate allocation scheme whose comprehensive score meets the preset conditions is selected as the second base station task sequence.

4. The method according to claim 2, characterized in that, Before obtaining the set of task base stations to be processed within the task distribution period, the method further includes: Initialize the first exponent matrix and the second exponent matrix. The elements in the i-th row and j-th column of the first exponent matrix and the second exponent matrix correspond to the base station pairs formed by the i-th base station and the j-th base station in a preset arrangement order. The initial value of each element is zero. For each pair of different base stations in the entire base station network, the total number of times that pair of base stations appear together in the same subset of base stations is counted, and the total number of times is recorded as the element value of the corresponding position in the first frequency matrix; The number of times each base station appears as an adjacent node in the optimal access path is counted, and the number of times the adjacent node appears is recorded as the element value of the corresponding position in the second frequency matrix; The first probability matrix is ​​constructed based on the ratio of each element value in the second exponent matrix to the corresponding element value in the first exponent matrix.

5. The method according to claim 4, characterized in that, The step of adjusting the frequencies of corresponding base station pairs in the first probability matrix based on the difference between the second base station task sequence and the first base station task sequence to obtain the second probability matrix includes: Obtain newly added base station pair connections in the second base station task sequence, wherein the newly added base station pair connections include base station pairs that are adjacent in the second base station task sequence but not adjacent in the first base station task sequence; Obtain the base station pair connections that have been removed from the first base station task sequence, wherein the removed base station pair connections include base station pairs that are adjacent in the first base station task sequence but not adjacent in the second base station task sequence; For each newly added base station pair connection, the record value of the corresponding element in the first exponent matrix is ​​added to obtain the adjusted second exponent matrix; For each removed base station connection, the recorded value of its corresponding element in the first exponent matrix is ​​reduced to obtain the adjusted second exponent matrix; Based on the adjusted second probability matrix and the first probability matrix, the connection probability of each base station pair is recalculated to obtain the second probability matrix.

6. The method according to claim 1, characterized in that, The step of generating a first base station task sequence based on the target probability matrix and the area of ​​responsibility of the staff includes: For each staff member, the intersection of their assigned area and the set of task base stations is determined, which serves as the set of base stations to be assigned to that staff member. Randomly select a base station from the set of base stations to be allocated as the starting base station of the task base station sequence; Remove the starting base station from the set of base stations to be allocated to obtain the remaining set of base stations; The starting base station is designated as the current base station; Repeat the following steps until the set of base stations to be allocated is empty to obtain the first base station task sequence: Calculate the distance from the current base station to each base station in the set of base stations to be allocated; From the target probability matrix, obtain the connection probability value between the current base station and each base station in the set of base stations to be allocated; For each base station in the set of base stations to be allocated, its distance to the current base station and the corresponding connection probability value are multiplied to obtain multiple comprehensive selection evaluation values; From the set of base stations to be allocated, select the base station with the smallest comprehensive selection evaluation value as the target base station; Remove the target base station from the set of base stations to be assigned and add it to the end of the worker's first base station task sequence; Update the target base station to the current base station.

7. The method according to claim 1, characterized in that, Before obtaining the set of task base stations to be processed within the task distribution period, the method further includes: For each subset of base stations, a two-dimensional coordinate matrix containing the coordinates of all base stations in that subset is constructed based on the latitude and longitude coordinates of all base stations in the subset. The coordinate transformation matrix of the base station subset is determined by extracting the extreme values ​​of the two-dimensional coordinate matrix in the longitude and latitude dimensions. Based on the coordinate transformation matrix, the two-dimensional coordinate matrix is ​​normalized to obtain a normalized coordinate matrix; The coordinates in the normalized coordinate matrix are used as the base station location information of the subset of base stations.

8. A task switching device, characterized in that, The device includes: The set acquisition module is used to acquire the set of task base stations that need to be processed within the task distribution period; The first selection module is used to select the first probability matrix as the target probability matrix when the set of task base stations includes base stations not covered by the second probability matrix. The second selection module is used to select the second probability matrix as the target probability matrix when all base stations in the task base station set are covered by the second probability matrix. The sequence generation module is used to generate a first base station task sequence based on the target probability matrix and the area of ​​responsibility of the staff. The task distribution module is used to distribute base station tasks corresponding to the first base station task sequence to the staff. The first probability matrix is ​​a base station pair connection probability matrix generated based on base station location information and base station subset sampling data. The sampling count value of each base station in the base station subset is greater than or equal to a preset threshold, and the sampling count value represents the total number of times the base station appears in all base station subsets. Each base station subset includes a central base station and multiple neighboring base stations. The central base station in each base station subset is selected from the set of base stations whose sampling count value reaches the global minimum value at that sampling time. The multiple neighboring base stations in each base station subset are the multiple base stations closest to the central base station of that subset. The connection probability value in the first probability matrix is ​​calculated based on the frequency with which each base station pair appears as an adjacent node in multiple optimal access paths determined by the base station subset sampling data. The second probability matrix is ​​obtained by adjusting the frequency of the corresponding base station pair in the first probability matrix according to the difference between the second base station task sequence and the first base station task sequence. The second base station task sequence is obtained by randomly perturbing the first base station task sequence corresponding to multiple staff members based on the execution feedback data. The execution feedback data is obtained by the multiple staff members executing their corresponding base station tasks.

9. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the task switching method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the task switching method as described in any one of claims 1-7.

11. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device performs the task switching method as described in any one of claims 1-7.

Images