Task offloading method, device, electronic device, and computer program product

Abstract

The present application provides a task offloading method, device, electronic device, and computer program product. The method includes: selecting a suspected stop point of a terminal from the terminal's historical movement trajectory; screening out the terminal's actual stop point from the suspected stop point according to a preset discrimination method, and obtaining the terminal's historical stop area; predicting the terminal's future stop area based on the terminal's historical stop area; obtaining an optimal offloading decision based on the relationship between the terminal's future stop area and base stations within a preset range from the future stop area, and unloading the task to be executed from the terminal based on the optimal offloading decision. The task offloading method provided in the embodiment of the present application can obtain an optimal offloading decision based on the relationship between the terminal's future stop area and the base stations surrounding the future stop area, and accurately control the unloading of the task to be executed on the terminal.

Description

Technical Field

[0001] The present application relates to the field of mobile edge computing technology, and specifically to a task offloading method, device, electronic device, and computer program product. Background Art

[0002] With the development of network and internet technologies, many applications have emerged, most of which are computationally intensive, such as those requiring facial recognition or voice processing, and mobile games. However, running computationally intensive applications on resource-constrained mobile user devices consumes a significant amount of device resources and battery life.

[0003] To address this issue, existing technologies can offload computational tasks from compute-intensive applications on mobile user devices to a mobile edge computing cloud server. Once the task is successfully completed, the mobile edge computing cloud server returns the program execution results to the mobile user device. This eliminates the need to run compute-intensive applications on the mobile user device, thereby reducing resource consumption on the mobile user device. Currently, when offloading a large number of computational tasks, multiple tasks are typically offloaded simultaneously to improve task offloading efficiency. This can easily lead to server congestion, resulting in delays and high energy consumption. Summary of the Invention

[0004] An embodiment of the present application provides a task offloading method to solve the technical problem that server congestion is likely to occur when multiple computing tasks are offloaded simultaneously, thereby causing delays and high energy consumption.

[0005] In a first aspect, an embodiment of the present application provides a task offloading method, comprising:

[0006] Select the suspected stop points of the terminal from the terminal's historical movement trajectory;

[0007] Filtering out the actual stay points of the terminal from the suspected stay points of the terminal according to a preset determination method, and obtaining the historical stay areas of the terminal;

[0008] predicting a future stay area of ​​a terminal based on a historical stay area of ​​the terminal;

[0009] An optimal offloading decision is obtained according to a relationship between the future stay area of ​​the terminal and base stations within a preset range from the future stay area, and the to-be-executed task is offloaded from the terminal according to the optimal offloading decision.

[0010] In one embodiment, selecting a suspected stop point of the terminal from the terminal's historical movement trajectory includes:

[0011] Calculating the actual movement distance and actual movement time of adjacent trajectory points in the historical movement trajectory of the terminal;

[0012] Adjacent trajectory points whose actual moving distance is less than a moving distance threshold and whose actual moving time is greater than a moving time threshold are selected as suspected stay points of the terminal.

[0013] In one embodiment, the filtering out the actual stay point of the terminal from the suspected stay points of the terminal according to a preset determination method and obtaining the historical stay area of ​​the terminal includes:

[0014] Determining correlation values ​​between adjacent suspected stay points among the suspected stay points of the terminal;

[0015] When the correlation value between the adjacent suspected stay points is greater than a preset threshold, the pair of adjacent suspected stay points is determined to be the actual stay points of the terminal;

[0016] Perform cluster analysis on the actual stay points of the terminal to obtain the historical stay areas of the terminal.

[0017] In one embodiment, the step of obtaining an optimal offloading decision based on a relationship between a future stay area of ​​the terminal and base stations within a preset range from the future stay area, and offloading the to-be-executed task from the terminal based on the optimal offloading decision, includes:

[0018] Determining whether a future stay area of ​​the terminal is in a coverage overlap area of ​​base stations within a preset range from the future stay area;

[0019] If the future stay area of ​​the terminal is not in the coverage overlap area of ​​the base station within the preset range from the future stay area, obtaining a first optimal offloading decision based on distance information between the future stay area of ​​the terminal and the base station within the preset range from the future stay area through a first optimal energy consumption model, and offloading the to-be-executed task from the terminal according to the first optimal offloading decision;

[0020] If the future stay area of ​​the terminal is in the coverage overlapping area of ​​the base station within the preset range of the future stay area, a second optimal unloading decision is obtained based on the distance information between the future stay area of ​​the terminal and the base station within the preset range of the future stay area and the base station influencing factors through a second optimal energy consumption model, and the task to be executed is unloaded from the terminal according to the second optimal unloading decision.

[0021] In one embodiment, obtaining a first optimal offloading decision based on distance information between the future stay area of ​​the terminal and the base station within a preset range from the future stay area using a first optimal energy consumption model, and offloading the to-be-executed task from the terminal according to the first optimal offloading decision, includes:

[0022] Inputting distance information between the future stay area of ​​the terminal and the base station within a preset range from the future stay area into the first optimal energy consumption model;

[0023] Obtaining the first optimal unloading decision by using the first optimal energy consumption model and a mutated firefly algorithm;

[0024] The to-be-executed task is offloaded from the terminal to the corresponding base station according to the first optimal offloading decision.

[0025] In one embodiment, obtaining a second optimal offloading decision based on the distance information between the future stay area of ​​the terminal and the base station within a preset range from the future stay area and the base station influencing factors using the second optimal energy consumption model, and offloading the to-be-executed task from the terminal according to the second optimal offloading decision, includes:

[0026] Inputting distance information between the future stay area of ​​the terminal and the base stations within a preset range from the future stay area and base station influencing factors into the second optimal energy consumption model;

[0027] Obtaining the second optimal unloading decision by using the second optimal energy consumption model and optimizing the simulated annealing intelligent algorithm;

[0028] The to-be-executed task is offloaded from the terminal to the corresponding base station according to the second optimal offloading decision.

[0029] In one embodiment, obtaining the second optimal unloading decision by using the second optimal energy consumption model and optimizing the simulated annealing intelligent algorithm includes:

[0030] Based on the distance information between the future stay area of ​​the terminal and the base stations within the preset range from the future stay area and the base station influencing factors, generating a pending offloading decision in a hierarchical manner according to preset conditions;

[0031] Obtaining an initial offloading decision from the pending offloading decisions using a strategy combining greedy and randomization;

[0032] Several initial uninstallation decisions are obtained by concurrent operation, and the second optimal uninstallation decision is selected from the several initial uninstallation decisions.

[0033] In a second aspect, an embodiment of the present application provides a task offloading device, comprising:

[0034] A suspected stop point selection module is used to select the suspected stop points of the terminal from the terminal's historical movement trajectory;

[0035] A historical stay area determination module is used to filter out the terminal's real stay points from the terminal's suspected stay points according to a preset determination method, and obtain the terminal's historical stay area;

[0036] A future stay area prediction module, configured to predict a future stay area of ​​a terminal based on a historical stay area of ​​the terminal;

[0037] The optimal offloading decision determining module is used to obtain an optimal offloading decision based on the relationship between the future stay area of ​​the terminal and the base stations within a preset range from the future stay area, and to offload the to-be-executed task from the terminal according to the optimal offloading decision.

[0038] In a third aspect, an embodiment of the present application provides an electronic device comprising a processor and a memory storing a computer program, wherein when the processor executes the program, the steps of the task offloading method described in the first aspect are implemented.

[0039] In a fourth aspect, an embodiment of the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the task offloading method described in the first aspect.

[0040] The task offloading method, device, electronic device, and computer program product provided in the embodiments of the present application first select the suspected stay point of the terminal from the historical movement trajectory of the terminal, then screen out the actual stay point of the terminal, and obtain the historical stay area of ​​the terminal, which can ensure the accuracy of the obtained historical stay area of ​​the terminal and the reliability of the subsequent prediction of the future stay area of ​​the terminal; then predict the future stay area of ​​the terminal based on the historical stay area of ​​the terminal, and then derive the optimal offloading decision based on the relationship between the future stay area of ​​the terminal and the base stations around the future stay area, and then unload the task to be executed from the terminal according to the optimal offloading decision, which can accurately control the unloading of the task to be executed from the terminal, and avoid the situation where server congestion is likely to occur when unloading multiple computing tasks at the same time, thereby causing delays and high energy consumption problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present application or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0042] Figure 1This is a flowchart of a task offloading method provided in an embodiment of the present application;

[0043] Figure 2 Shows that there may be overlapping coverage areas between base stations;

[0044] Figure 3 This is a module diagram of a task offloading device provided in an embodiment of the present application;

[0045] Figure 4 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions in this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.

[0047] Figure 1 A flowchart of the task offloading method provided in an embodiment of the present application.

[0048] Reference Figure 1 , an embodiment of the present application provides a task offloading method, which may include the steps of:

[0049] S1: Select the terminal's suspected stop points from the terminal's historical movement trajectory.

[0050] S2: Filter out the actual stay point of the terminal from the suspected stay points of the terminal according to a preset discrimination method, and obtain the historical stay area of ​​the terminal.

[0051] S3: Predict the terminal's future stay area based on the terminal's historical stay area.

[0052] S4: deriving an optimal offloading decision based on a relationship between the terminal's future stay area and base stations within a preset range from the future stay area, and offloading the to-be-executed task from the terminal based on the optimal offloading decision.

[0053] It should be noted that the execution subject of the task offloading method provided by the present invention may be a terminal-side device, such as a user equipment.

[0054] Before the terminal side device executes steps S1-S4, the terminal side device will first execute S01:

[0055] Get the terminal's historical movement trajectory.

[0056] Specifically, the historical movement trajectory of the terminal can be obtained from the background data center. The information contained in the historical movement trajectory of the terminal may include the location information where the terminal stayed during the past movement, the time information at each location, the number of times the terminal stayed at each location, etc.

[0057] In step S1, the terminal side device selects the suspected stop point of the terminal from the historical movement trajectory of the terminal.

[0058] For example, the suspected stay point of the terminal can be obtained by comparing the information included in the historical movement trajectory of the terminal and the preset suspected threshold value corresponding to each piece of information, thereby obtaining a comprehensive evaluation.

[0059] In step S2, the terminal side device will filter out the real stay point of the terminal from the suspected stay points of the terminal according to a preset determination method, and obtain the historical stay area of ​​the terminal.

[0060] Specifically, after the actual stay points of the terminal are screened out from the suspected stay points of the terminal by a preset discrimination method, a cluster analysis can be performed on the actual stay points of the terminal to obtain the historical stay areas of the terminal.

[0061] In step S3, the terminal side device predicts the terminal's future stay area based on the terminal's historical stay area.

[0062] For example, the terminal's future stay area may be predicted based on the distribution pattern of each real stay point in the terminal's historical stay area.

[0063] In step S4, the terminal side device will obtain an optimal offloading decision based on the relationship between the terminal's future stay area and the base stations within a preset range from the future stay area, and offload the to-be-executed task from the terminal according to the optimal offloading decision.

[0064] Specifically, the number of base stations within the preset range of the future stay area can be one or more than two. When the number of base stations within the preset range of the future stay area is one, the optimal unloading decision may include the time point information of unloading the tasks to be executed from the terminal to the base station, which is conducive to the base station to process the tasks to be executed in an orderly manner and avoid congestion.

[0065] When the number of base stations within the preset range of the future stay area is more than two, the coverage areas of these base stations may have overlapping coverage areas and non-overlapping coverage areas. The relationship between the future stay area of ​​the terminal and the base stations within the preset range of the future stay area may be that the future stay area of ​​the terminal is in the overlapping coverage area or in the non-overlapping coverage area. The terminal side device will derive the corresponding optimal unloading decision based on the relationship between the future stay area of ​​the terminal and the base stations within the preset range of the future stay area, so as to determine which base station is the best solution to offload the terminal's pending tasks to through the optimal unloading decision.

[0066] The task offloading method provided in the embodiment of the present application first selects the suspected stay points of the terminal from the historical movement trajectory of the terminal, then filters out the real stay points of the terminal, and obtains the historical stay area of ​​the terminal, which can ensure the accuracy of the obtained historical stay area of ​​the terminal and the reliability of the subsequent prediction of the future stay area of ​​the terminal; then the future stay area of ​​the terminal is predicted based on the historical stay area of ​​the terminal, and then the optimal offloading decision is obtained based on the relationship between the future stay area of ​​the terminal and the base stations within a preset range from the future stay area. Then, the task to be executed is unloaded from the terminal according to the optimal offloading decision, which can accurately control the unloading of the task to be executed from the terminal, and avoid the situation where server congestion is likely to occur when multiple computing tasks are unloaded at the same time, thereby causing delays and high energy consumption problems.

[0067] In one embodiment, step S1 may include:

[0068] Preset the moving distance threshold and moving time threshold between adjacent trajectory points in the historical movement trajectory of the terminal;

[0069] Calculate the actual movement distance and actual movement time of adjacent trajectory points in the terminal's historical movement trajectory;

[0070] Each pair of adjacent trajectory points whose real moving distance is less than the moving distance threshold and whose real moving time is greater than the moving time threshold is selected as the suspected stop point of the terminal.

[0071] By comparing the actual moving distance and actual moving time of adjacent trajectory points in the historical movement trajectory with the moving distance threshold and moving time threshold, a unified judgment standard is used to obtain the suspected stay point of the terminal, making the obtained suspected stay point of the terminal more accurate and providing reliable data for subsequent steps.

[0072] In one embodiment, step S2 may include:

[0073] Calculating correlation values ​​between adjacent suspected stay points among the suspected stay points of the terminal;

[0074] When the correlation between adjacent suspected stay points is greater than a preset threshold, the adjacent suspected stay points are determined to be the actual stay points of the terminal;

[0075] Perform cluster analysis on the actual stay points of the terminal to obtain the terminal's historical stay areas.

[0076] The actual stay point of the terminal is obtained based on the correlation between adjacent suspected stay points among the suspected stay points of the terminal. Then, through cluster analysis of the actual stay points, the accurate historical stay area of ​​the terminal can be effectively obtained, making the subsequent prediction of the future stay area of ​​the terminal more refined.

[0077] In one embodiment, the correlation value between adjacent suspected stay points of the terminal is determined, which can be calculated using formula (1) and formula (2):

[0078]

[0079]

[0080] Among them, SP i represents the quantitative movement information of the i-th suspected stay point, d i represents the moving distance of the i-th suspected stop point, represents the average moving distance of the suspected stay points; n represents the number of suspected stay points, v i represents the moving speed of the i-th suspected stop point, represents the average moving speed of the suspected stop point; Δt represents the moving time between two adjacent suspected stop points. represents the average moving time of the suspected stay point, Q(sp i ,sp i-1 ) represents the correlation value between two adjacent suspected stay points; Cov(sp i ,sp i-1 ) represents the covariance between two adjacent suspected stay points; σ 2 (sp i ) represents the variance of the quantized movement information of the i-th suspected stay point.

[0081] In addition, in addition to calculating the movement quantization information of a single suspected stay point through equations (1) and (2) to obtain the correlation between adjacent suspected stay points, the average movement quantization information of all suspected stay points can also be calculated to avoid the isolated point problem that may occur when only the movement quantization information of a single suspected stay trajectory point is calculated, thereby improving the accuracy of subsequent screening of true stay points.

[0082] In one embodiment, cluster analysis is performed on the actual stay points of the terminal to obtain the terminal's historical stay areas. A density-based clustering algorithm can be used to cluster the actual stay points of all terminals to obtain the historical stay areas, and the trajectory point sequence in the original trajectory dataset is converted into a stay area sequence. This step can refer to existing related technologies.

[0083] For example, the DBSCAN algorithm is used to cluster the real stay points of the terminal. First, a real stay point is selected and all the real stay points with a distance less than or equal to a radius r from the real stay point are found; then, if the number of real stay points with a distance less than the radius r is greater than the minimum number of points, the real stay point is marked as a core point and assigned a new cluster label; then all neighbor points within the radius r of the real stay point are visited. If they have not been assigned a cluster, the newly created cluster label is assigned to them. If they are core points, their neighbors are visited in turn, and so on, so that the cluster gradually increases until there are no core points within the radius r of the cluster; then another real stay point that has not been visited is selected and the same process is repeated until all the real stay points of the terminal are marked, thereby dividing the real stay points into clusters and obtaining the historical stay area corresponding to the real stay point.

[0084] In one embodiment, step S3 can combine the information of the terminal's historical stay area and the terminal's historical positioning data to determine the terminal's location transfer probability, and then compare the sizes of various transfer probabilities, so as to predict the terminal's future stay area based on the historical positioning data corresponding to the maximum transfer probability.

[0085] Among them, the historical positioning data may include any one or more of historical GPS positioning data, historical base station positioning data, historical WIFI positioning data, historical inertial navigation positioning data, and historical visual positioning data. If there are multiple types of historical positioning data obtained, different types of historical positioning data can be directly clustered according to a clustering algorithm to obtain clustering calculation results, and then the clustering calculation results corresponding to different types of historical positioning data can be calculated to obtain different activity areas. After the location data is predicted based on the activity areas corresponding to different types of historical positioning data, the location data is averaged to obtain better location data. In this embodiment, after averaging, the diversity and difference of the data can be increased, and data overfitting can be avoided, thereby making the information of the predicted terminal's future stay area more accurate.

[0086] The detailed implementation process of calculating the transfer probability and predicting the future stay area can be achieved by referring to existing related technologies.

[0087] In one embodiment, step S4 may include:

[0088] Determining whether the future stay area of ​​the terminal is in a coverage overlap area of ​​base stations within a preset range from the future stay area;

[0089] If the future stay area of ​​the terminal is not in the coverage overlap area of ​​the base station within the preset range from the future stay area, obtaining a first optimal offloading decision based on distance information between the future stay area of ​​the terminal and the base station within the preset range from the future stay area through a first optimal energy consumption model, and offloading the to-be-executed task from the terminal according to the first optimal offloading decision;

[0090] If the future stay area of ​​the terminal is in the coverage overlapping area of ​​the base station within the preset range of the future stay area, the second optimal unloading decision is obtained based on the distance information between the future stay area of ​​the terminal and the base station within the preset range of the future stay area and the base station influencing factors through the second optimal energy consumption model, and the task to be executed is unloaded from the terminal according to the second optimal unloading decision.

[0091] The optimal offloading decision is made based on the relationship between the terminal's future stay area and the base stations within a preset range from the future stay area. Then, the tasks to be executed are offloaded from the terminal based on the optimal offloading decision. This can accurately control the offloading of tasks to be executed from the terminal, avoiding server congestion that is prone to occur when unloading multiple computing tasks at the same time, thereby causing delays and high energy consumption problems.

[0092] In one embodiment, when determining whether the future stay area of ​​the mobile user equipment is within the coverage overlap area of ​​base stations surrounding the future stay area, the following method may be used:

[0093] Among the base stations around the future stay area, all base stations whose distances between the future stay area of ​​the mobile user equipment and the base stations around the future stay area are less than a preset threshold are selected, and then the Thiessen polygons of each base station are divided with each base station as the center (the Thiessen polygons can represent the coverage area of ​​the base station), and it is determined whether the Thiessen polygons have overlapping areas. If so, it is determined that the future stay area of ​​the mobile user equipment is in the coverage overlapping area of ​​the base stations around the future stay area; if not, it is determined that the future stay area of ​​the mobile user equipment is not in the coverage overlapping area of ​​the base stations around the future stay area.

[0094] Among them, when dividing the Thiessen polygons of each base station, you can first build a triangle network, then traverse each vertex, calculate the center of the circumscribed circle adjacent to it; finally, connect the centers of all circumscribed circles to obtain the Thiessen polygons of each base station.

[0095] In actual applications, since the antennas used in base stations are often omnidirectional or tridirectional, the coverage of the base station is circular when viewed from the plane, not in the ideal honeycomb shape, which easily causes overlapping coverage of the base station (such as Figure 2 As shown in the figure, in this case, if the predicted stay area happens to be in the coverage overlap area of ​​multiple base stations (such as areas A, B, C, D, and E), then the current data transmission rate is not only affected by the distance between the future stay area of ​​the mobile user equipment and the base station, but also by factors such as the base station received signal strength, the amount of task data processed by the base station, and the maximum allowable delay of the base station. The present invention first determines whether the future stay area of ​​the mobile user equipment is in the coverage overlap area of ​​the base stations surrounding the future stay area, and then derives different optimal task offloading strategies based on different situations. This can effectively obtain a more accurate task offloading strategy based on local conditions, thereby improving task offloading efficiency.

[0096] On the one hand, when the future staying area of ​​the terminal is not in the coverage overlapping area of ​​the base station within the preset range of the future staying area, the distance between the future staying area of ​​the terminal and the base station within the preset range of the future staying area and the data transmission rate of each base station can be calculated first, and then the first optimal unloading decision can be obtained through the first optimal energy consumption model.

[0097] In one embodiment, obtaining a first optimal offloading decision based on distance information between a future stay area of ​​the terminal and a base station within a preset range from the future stay area using a first optimal energy consumption model, and offloading the to-be-executed task from the terminal according to the first optimal offloading decision may include the following steps:

[0098] Inputting distance information between the terminal's future stay area and a base station within a preset range from the future stay area into a first optimal energy consumption model;

[0099] The first optimal unloading decision is obtained by using the first optimal energy consumption model and the mutant firefly algorithm;

[0100] The to-be-executed task is offloaded from the terminal to the corresponding base station according to the first optimal offloading decision.

[0101] Specifically, the distance information between the terminal's future stay area and the base stations within the preset range from the future stay area may include the actual distance between the terminal's future stay area and the base stations within the preset range from the future stay area and the data transmission rate of each base station.

[0102] In one embodiment, the actual distance between the future stay area of ​​the terminal and a base station within a preset range from the future stay area can be calculated using formula (3):

[0103]

[0104] Among them, d 基站 represents the distance between the terminal's future stay area and the base station within the preset range from the future stay area, R is the radius of the earth, The future stay area of ​​the terminal and the longitude and latitude of the base stations within a preset range from the future stay area.

[0105] After calculating the distance between the terminal's future stay area and the base station within a preset range from the future stay area, the data transmission rate can be further calculated using formula (4).

[0106]

[0107] Where w is the system bandwidth, σ is the noise power, p is the terminal transmit power, g0(d0 / d) 4 Denoted as h, represents the channel power gain, g0 is the path loss constant, and d0 is the reference distance.

[0108] By the above method, the actual distance value between the future stay area of ​​the terminal and the base stations within the preset range from the future stay area and the data transmission rate of each base station can be accurately calculated.

[0109] The information included in the first optimal energy consumption model may be:

[0110]

[0111]

[0112]

[0113]

[0114] Where n is the number of terminals, t i is the actual execution time of the task, r i Indicates the data transmission rate, p i represents the transmit power of terminal i, d i represents the moving distance of the i-th suspected stop point, δ 0 represents the energy consumption of a CPU cycle when the cloud server executes a task, C represents the computing power of the cloud server, and x i Indicates whether base station i is available, where x i =1 means it is available. i =0 means unavailable.

[0115] In one embodiment, regarding the use of a mutated firefly algorithm through a first optimal energy consumption model to solve the energy consumption minimization to obtain a first offloading decision, specifically, the position of the firefly corresponds to a solution vector of the optimization problem, namely the first optimal offloading decision, and each component of this vector is 1 or 0, 1 means that the terminal's task is offloaded to the cloud server for execution, and 0 means that the terminal's task is not offloaded to the cloud server for execution; wherein, the number of dimensions of the position vector corresponds to the number of tasks, and the objective function value corresponds to the energy consumption value, and fireflies with lower energy consumption values ​​attract fireflies with higher energy consumption values ​​to move toward them.

[0116] When applying the mutated firefly algorithm to solve the first offloading decision, in order to make the first offloading decision feasible, during the iteration process, if the local execution time of the task corresponding to a component is greater than the delay constraint, that is, executing this task on the terminal does not meet the delay constraint, then the priority value of this component is set to the maximum value 1 and is unloaded first; then a greedy strategy is adopted to examine whether the total computing power required by the current task meets the computing power constraint of the cloud server in order of priority from high to low. If so, this task can be unloaded at this base station. If not, the next task is examined in sequence to see if it can be unloaded at this base station until the end.

[0117] Finally, the corresponding component value of the offloading task at the corresponding base station is 1, otherwise it is 0, and the first optimal offloading decision is obtained comprehensively.

[0118] On the other hand, when the future stay area of ​​the terminal is in the coverage overlapping area of ​​the base station within the preset range of the future stay area, the terminal side device will obtain a second optimal unloading decision based on the distance information between the future stay area of ​​the terminal and the base station within the preset range of the future stay area and the base station influencing factors through the second optimal energy consumption model, and unload the task to be executed from the terminal according to the second optimal unloading decision.

[0119] Specifically, the base station influencing factors may include one or more of the base station's received signal strength, the amount of task data processed by the base station, and the base station's maximum allowable delay.

[0120] In one embodiment, obtaining a second optimal offloading decision based on the distance information between the future stay area of ​​the terminal and a base station within a preset range from the future stay area and a base station influencing factor using a second optimal energy consumption model, and offloading the to-be-executed task from the terminal according to the second optimal offloading decision may include the following steps:

[0121] Inputting distance information between the terminal's future stay area and a base station within a preset range from the future stay area and base station influencing factors into a second optimal energy consumption model;

[0122] The second optimal unloading decision is obtained by using the optimized simulated annealing intelligent algorithm through the second optimal energy consumption model;

[0123] The task to be executed is offloaded from the terminal to the corresponding base station according to the second optimal offloading decision.

[0124] The information included in the second optimal energy consumption model may be:

[0125]

[0126]

[0127] ......

[0129]

[0130] x i =1or 0,

[0131] Where m represents the number of base stations whose distances between the terminal's future stay area and the base stations within the preset range from the future stay area are less than a preset threshold, s i represents the utility estimate of base station i; q represents the number of base station influencing factors that affect the transmission rate. For example, in this embodiment, the base station influencing factors include the base station's received signal strength, the amount of task data processed by the base station, and the base station's maximum allowable delay. Then q = 3; Limit j (j=1,2,…,q) represents the index constraints corresponding to the influencing factors of each base station; c qi represents the energy consumption of the i-th base station under the influence of the q-th base station, x i Indicates whether the base station is available, where x i =1 means it is available. i =0 means unavailable.

[0132] In one embodiment, obtaining the second optimal unloading decision by optimizing the simulated annealing intelligent algorithm (SA) through the second optimal energy consumption model may include the following steps:

[0133] Based on the distance information between the terminal's future stay area and a base station within a preset range from the future stay area and base station influencing factors, a pending offloading decision is generated hierarchically according to preset conditions;

[0134] A strategy combining greedy and randomization is used to obtain the initial offloading decision from the pending offloading decisions;

[0135] Running the programs concurrently to obtain several initial offloading decisions, and selecting the second best offloading decision from among the several initial offloading decisions;

[0136] The task to be executed is offloaded from the terminal to the corresponding base station according to the second optimal offloading decision.

[0137] Exemplarily, the following describes a process of obtaining the second optimal unloading decision using an optimized simulated annealing intelligent algorithm:

[0138] Step A, configure initialization parameters:

[0139] Initial temperature T (sufficiently large): for example 100000;

[0140] The lower limit of the temperature at which the algorithm ends Tmin: for example, 1e-8;

[0141] The number of disturbance states K at the same temperature: for example, 100;

[0142] Temperature drop rate delta (less than 1 but close to 1): for example 0.99;

[0143] The number of base stations around the terminal whose distance is less than the preset threshold: m;

[0144] Service performance indicator consumption of base station X: C;

[0145] The utility score of base station X is s i : [s1, s2, ... s m ].

[0146] Step B, optimization solution:

[0147] B-1: Using parameters X, C, and Limit, initialize K groups of initial solutions (initial offloading decisions) a[K][m] that meet the Limit constraints. The generation of these initial solutions is not completely randomized. To improve the efficiency of searching for the global optimal solution, this embodiment uses a strategy that combines a small amount of greedy and a large amount of randomization to generate initial solutions. Based on actual base station scenarios, for example, a greedy strategy generates three groups of initial solutions. The remaining K-3 groups of initial solutions are generated randomly.

[0148] Among them, there are two greedy strategies: the first one is to sort the base stations X from large to small according to their utility scores s, and select a group of initial solutions formed by submodules that meet the Limit constraints in sequence; the second one is to sort the base stations X from large to small according to their utility score density ρ (ρ = utility score s / consumption C of the corresponding service performance indicator), and select L groups of initial solutions formed by submodules that meet the Limit constraints in sequence (where L = the number of base station influencing factors, which in this embodiment can include three: the base station's received signal strength, the amount of task data processed by the base station, and the base station's maximum allowable delay. Note: If consumption C is 0, then ρ = utility score s / [a number close to 0, which can be 0.01 here]).

[0149] Each solution is in the form [xi1 , x i2 ,..., x im where x ij = 0 or 1. For example, the initial solution [1, 0, 1, 0…0, 0] represents selecting the 1st and 3rd tasks in the terminal for offloading to this base station;

[0150] B-2: If the initial temperature T has not reached the temperature lower bound Tmin, repeat the following B-3 to B-6.

[0151] B-3: At the current T state, loop K times. Each time, use the current a[i] (i = 1, 2, 3, ……, K) as a feasible solution to generate a pending new solution b i : The way to generate a new solution is not to generate it in a fixed pattern in the standard SA; to reduce the probability of the solution process falling into a local optimal solution, this embodiment proposes to generate a pending solution (pending offloading decision) within the neighborhood of the current feasible solution according to the current temperature t value. Specifically, the levels are divided into three levels: high, medium, and low:

[0152] High-level range: t > 2 / 3 * T; randomly reverse 3 elements of the a[i] array in the neighborhood;

[0153] Medium-level range: 1 / 3 * T < t <= 2 / 3 * T; randomly reverse 2 elements of the a[i] array in the neighborhood;

[0154] Low-level range: t <= 1 / 3 * T; randomly reverse 1 element of the a[i] array in the neighborhood.

[0155] Here, to ensure that only small changes occur in the neighborhood of the current solution, the maximum number of reversals is controlled at about 10% of the total number of base stations; the meaning of random reversal is: reverse one element of the a[i] array, then for the pending new solution b i [i][j] = 1 then reverse the value at position j to 0, otherwise to 1; for example, if the randomly reversed position of the current solution is the hot song recall module and its value is 1, then the pending new solution will not select this base station, otherwise it will be considered; at the same time, calculate the utility value s corresponding to the current feasible solution a[i] i (note that it becomes negative), which can be specifically calculated by Equation (5):

[0156]

[0157] where s i represents the utility value of base station i, m represents the number of base stations within the preset threshold distance around the terminal, x i characterizes whether the base station is available, where when x i = 1 it means available, and when x i = 0 it means unavailable.

[0158] B-4: Determine the new solution b i Check if the Limit constraint is met. If so, proceed to B-5.

[0159] B-5: Calculate the new solution b i The corresponding recommendation utility value s i (Note that it becomes a negative number).

[0160] B-6: Determine the new solution b i Acceptance: Using the Metropolis criterion of SA, which accepts bad solutions with a certain probability (making the utility value smaller), so that the algorithm has the ability to get rid of local extreme values ​​and avoid premature convergence, accept the new solution b i The criteria are:

[0161]

[0162] df=E(x new )-E(x) (6)

[0163] Where t is the current temperature, exp is the natural exponential, and df is the new solution b of the i-th inner loop at the current temperature t. i With the current solution a i The difference in utility value; the smaller the temperature t, the smaller the cooling probability Y; the higher the temperature t, the greater the cooling probability Y. The cooling probability Y reflects the probability of accepting the new solution b i As the new current solution a i If the probability degree is met, then b i Replace a i Become the current solution; otherwise, discard the new solution b i , current a i constant.

[0164] B-7: Update the current T state according to t = temperature drop rate delta*T.

[0165] B-8: Determine whether t reaches the lower limit Tmin of the temperature at which the SA algorithm ends; if not, execute B-3; otherwise, execute B-9.

[0166] B-9: Traverse the last a[i] (i=1,2,3,…,K) and select the estimated maximum recommendation utility a[i] vector (elements with 1 in a[i]).

[0167] Since SA theoretically converges to the global optimal solution of the objective function with a probability of 1 (i.e., convergence in a statistical sense, not deterministic convergence), this embodiment runs the above-mentioned solution process concurrently. Among the N optimal solutions (optimal initial offloading decisions) obtained under the concurrency number N, the final optimal solution, i.e., the second optimal offloading decision, is selected, and then the task to be executed is offloaded from the terminal to the corresponding base station according to the second optimal offloading decision.

[0168] The following describes a task offloading device provided in an embodiment of the present application. The task offloading device described below and the task offloading method described above can refer to each other.

[0169] like Figure 3 As shown, an embodiment of the present application provides a task offloading device, comprising:

[0170] A suspected stay point selection module 310 is configured to select a suspected stay point of the terminal from the terminal's historical movement trajectory;

[0171] A historical stay area determination module 320 is configured to filter out the terminal's actual stay points from the terminal's suspected stay points according to a preset determination method, and obtain the terminal's historical stay areas;

[0172] A future stay area prediction module 330 is configured to predict a future stay area of ​​a terminal based on the terminal's historical stay areas;

[0173] The optimal offloading decision determining module 340 is configured to obtain an optimal offloading decision based on the relationship between the terminal's future stay area and base stations within a preset range from the future stay area, and offload the to-be-executed task from the terminal based on the optimal offloading decision.

[0174] Furthermore, the suspected stay point selection module 310 is specifically configured to:

[0175] Calculate the real moving distance and real moving time of adjacent trajectory points in the historical movement trajectory of the terminal, and select the adjacent trajectory points whose real moving distance is less than the moving distance threshold and whose real moving time is greater than the moving time threshold as the suspected stay point of the terminal.

[0176] Furthermore, the historical stay area determination module 320 includes:

[0177] a correlation calculation submodule, configured to determine correlation values ​​between adjacent suspected stay points among the suspected stay points of the terminal;

[0178] a real stay point determination submodule, configured to determine, when a correlation value between the adjacent suspected stay points is greater than a preset threshold, that the pair of adjacent suspected stay points is the real stay point of the terminal;

[0179] The real stay point clustering submodule is used to perform cluster analysis on the real stay points of the terminal to obtain the historical stay areas of the terminal.

[0180] Furthermore, the optimal unloading decision determination module 340 includes:

[0181] A terminal-base station relationship determination submodule, configured to determine whether the terminal's future stay area is within a coverage overlap area of ​​base stations within a preset range from the future stay area;

[0182] a first task offloading submodule, configured to obtain, by using a first optimal energy consumption model and based on distance information between the future stay area of ​​the terminal and the base station within the preset range from the future stay area, a first optimal offloading decision if the future stay area of ​​the terminal is not in a coverage overlap area of ​​the base station within the preset range from the future stay area, and to offload the to-be-executed task from the terminal according to the first optimal offloading decision;

[0183] The second task unloading submodule is used to obtain a second optimal unloading decision based on the distance information between the future stay area of ​​the terminal and the base station within the preset range of the future stay area and the base station influencing factors through a second optimal energy consumption model if the future stay area of ​​the terminal is in the coverage overlapping area of ​​the base station within the preset range of the future stay area, and unload the task to be executed from the terminal according to the second optimal unloading decision.

[0184] Furthermore, the first task offloading submodule includes:

[0185] a distance information input submodule, configured to input distance information between the future stay area of ​​the terminal and the base station within a preset range from the future stay area into the first optimal energy consumption model;

[0186] a variant firefly algorithm calculation submodule, configured to obtain the first optimal unloading decision by using the variant firefly algorithm through the first optimal energy consumption model;

[0187] The first offloading submodule is configured to offload the to-be-executed task from the terminal to a corresponding base station according to the first optimal offloading decision.

[0188] Furthermore, the second task offloading submodule includes:

[0189] a distance information and base station influencing factor input submodule, configured to input the distance information between the future stay area of ​​the terminal and the base stations within a preset range from the future stay area and the base station influencing factors into the second optimal energy consumption model;

[0190] An optimized simulated annealing intelligent algorithm calculation submodule, configured to obtain the second optimal unloading decision by using the optimized simulated annealing intelligent algorithm through the second optimal energy consumption model;

[0191] The second offloading submodule is configured to offload the to-be-executed task from the terminal to the corresponding base station according to the second optimal offloading decision.

[0192] Furthermore, the optimized simulated annealing intelligent algorithm calculation submodule includes:

[0193] A pending offloading decision obtaining submodule is configured to generate a pending offloading decision hierarchically according to preset conditions based on distance information between the future stay area of ​​the terminal and the base stations within a preset range from the future stay area and base station influencing factors;

[0194] An initial unloading decision obtaining submodule is used to obtain an initial unloading decision from the pending unloading decisions by utilizing a strategy combining greedy and randomization;

[0195] The second optimal decision obtaining submodule is used to concurrently run to obtain several of the initial uninstallation decisions, and select the second optimal uninstallation decision from the several initial uninstallation decisions.

[0196] Figure 4 An example of a physical structure diagram of an electronic device is shown below. Figure 4 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 may call a computer program in the memory 830 to execute the steps of the task offloading method, for example, including:

[0197] Select the suspected stop points of the terminal from the terminal's historical movement trajectory;

[0198] The terminal's actual stay points are screened out from the terminal's suspected stay points according to a preset identification method, and the terminal's historical stay areas are obtained;

[0199] Predicting the terminal's future stay area based on the terminal's historical stay area;

[0200] An optimal offloading decision is obtained according to a relationship between a future stay area of ​​the terminal and base stations within a preset range from the future stay area, and the task to be executed is offloaded from the terminal according to the optimal offloading decision.

[0201] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0202] On the other hand, embodiments of the present application further provide a computer program product, comprising a computer program. The computer program may be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the steps of the task offloading method provided in each of the above embodiments, for example, including:

[0203] Select the suspected stop points of the terminal from the terminal's historical movement trajectory;

[0204] The terminal's actual stay points are screened out from the terminal's suspected stay points according to a preset identification method, and the terminal's historical stay areas are obtained;

[0205] Predicting the terminal's future stay area based on the terminal's historical stay area;

[0206] An optimal offloading decision is obtained according to a relationship between a future stay area of ​​the terminal and base stations within a preset range from the future stay area, and the task to be executed is offloaded from the terminal according to the optimal offloading decision.

[0207] On the other hand, an embodiment of the present application further provides a processor-readable storage medium, wherein the processor-readable storage medium stores a computer program, wherein the computer program is configured to cause a processor to execute the steps of the methods provided in the above embodiments, for example, including:

[0208] Select the suspected stop points of the terminal from the terminal's historical movement trajectory;

[0209] The terminal's actual stay points are screened out from the terminal's suspected stay points according to a preset identification method, and the terminal's historical stay areas are obtained;

[0210] Predicting the terminal's future stay area based on the terminal's historical stay area;

[0211] An optimal offloading decision is obtained according to a relationship between a future stay area of ​​the terminal and base stations within a preset range from the future stay area, and the task to be executed is offloaded from the terminal according to the optimal offloading decision.

[0212] The processor-readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO)), optical storage (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (such as ROMs, EPROMs, EEPROMs, non-volatile memories (NANDFLASH), solid-state drives (SSDs)), etc.

[0213] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0214] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0215] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A task offloading method, characterized in that: include: Select the suspected stop points of the terminal from the terminal's historical movement trajectory; Filtering out the actual stay points of the terminal from the suspected stay points of the terminal according to a preset determination method, and obtaining the historical stay areas of the terminal; predicting a future stay area of ​​a terminal based on a historical stay area of ​​the terminal; Determining an optimal offloading decision based on a relationship between a future stay area of ​​the terminal and base stations within a preset range from the future stay area, and offloading the to-be-executed task from the terminal based on the optimal offloading decision; The step of obtaining an optimal offloading decision based on a relationship between a future stay area of ​​the terminal and base stations within a preset range from the future stay area, and offloading the to-be-executed task from the terminal based on the optimal offloading decision, includes: Determining whether a future stay area of ​​the terminal is in a coverage overlap area of ​​base stations within a preset range from the future stay area; If the future stay area of ​​the terminal is not in the coverage overlap area of ​​the base station within the preset range from the future stay area, obtaining a first optimal offloading decision based on distance information between the future stay area of ​​the terminal and the base station within the preset range from the future stay area through a first optimal energy consumption model, and offloading the to-be-executed task from the terminal according to the first optimal offloading decision; If the future stay area of ​​the terminal is in the coverage overlap area of ​​the base station within the preset range from the future stay area, obtaining a second optimal offloading decision based on the distance information between the future stay area of ​​the terminal and the base station within the preset range from the future stay area and the base station influencing factors through a second optimal energy consumption model, and offloading the to-be-executed task from the terminal according to the second optimal offloading decision; The step of obtaining a second optimal offloading decision based on distance information between the future stay area of ​​the terminal and the base station within a preset range from the future stay area and a base station influencing factor using a second optimal energy consumption model, and offloading the task to be executed from the terminal according to the second optimal offloading decision, includes: Inputting distance information between the future stay area of ​​the terminal and the base stations within a preset range from the future stay area and base station influencing factors into the second optimal energy consumption model; Obtaining the second optimal unloading decision by using the second optimal energy consumption model and optimizing the simulated annealing intelligent algorithm; offloading the to-be-executed task from the terminal to the corresponding base station according to the second optimal offloading decision; The information contained in the second optimal energy consumption model is: Wherein, m represents the number of base stations whose distances between the future stay area of ​​the terminal and the base stations within the preset range from the future stay area are less than a preset threshold, s i represents the utility score of base station i; q represents the number of base station influencing factors that affect the transmission rate, including the base station's received signal strength, the amount of task data processed by the base station, and the base station's maximum allowable delay; Limit j (j=1, 2, ..., q) represents the index constraint corresponding to each of the base station influencing factors; c qi represents the energy consumption of the i-th base station under the influence of the q-th base station, x i Indicates whether base station i is available.

2. The task offloading method according to claim 1, characterized in that: The selecting of the suspected stopping point of the terminal from the historical movement trajectory of the terminal includes: Calculating the actual movement distance and actual movement time of adjacent trajectory points in the historical movement trajectory of the terminal; Adjacent trajectory points whose actual moving distance is less than a moving distance threshold and whose actual moving time is greater than a moving time threshold are selected as suspected stay points of the terminal.

3. The task offloading method according to claim 2, characterized in that: The method of screening out the actual stay point of the terminal from the suspected stay points of the terminal according to a preset determination method and obtaining the historical stay area of ​​the terminal includes: Determining correlation values ​​between adjacent suspected stay points among the suspected stay points of the terminal; When the correlation value between the adjacent suspected stay points is greater than a preset threshold, the pair of adjacent suspected stay points is determined to be the actual stay points of the terminal; Perform cluster analysis on the actual stay points of the terminal to obtain the historical stay areas of the terminal.

4. The task offloading method according to claim 1, wherein: The step of obtaining a first optimal offloading decision based on distance information between the future stay area of ​​the terminal and the base station within a preset range from the future stay area by using a first optimal energy consumption model, and offloading the task to be executed from the terminal according to the first optimal offloading decision, includes: Inputting distance information between the future stay area of ​​the terminal and the base station within a preset range from the future stay area into the first optimal energy consumption model; Obtaining the first optimal unloading decision by using the first optimal energy consumption model and a mutated firefly algorithm; The to-be-executed task is offloaded from the terminal to the corresponding base station according to the first optimal offloading decision.

5. The task offloading method according to claim 1, characterized in that: The step of obtaining the second optimal unloading decision by using the second optimal energy consumption model and optimizing the simulated annealing intelligent algorithm includes: Based on the distance information between the future stay area of ​​the terminal and the base stations within the preset range from the future stay area and the base station influencing factors, generating a pending offloading decision in a hierarchical manner according to preset conditions; Obtaining an initial offloading decision from the pending offloading decisions using a strategy combining greedy and randomization; Several initial uninstallation decisions are obtained by concurrent operation, and the second optimal uninstallation decision is selected from the several initial uninstallation decisions.

6. A task offloading device, characterized in that: include: A suspected stop point selection module is used to select the suspected stop points of the terminal from the terminal's historical movement trajectory; A historical stay area determination module is used to filter out the terminal's real stay points from the terminal's suspected stay points according to a preset determination method, and obtain the terminal's historical stay area; A future stay area prediction module, configured to predict a future stay area of ​​a terminal based on a historical stay area of ​​the terminal; an optimal offloading decision determining module, configured to derive an optimal offloading decision based on a relationship between a future stay area of ​​the terminal and base stations within a preset range from the future stay area, and offload the to-be-executed task from the terminal based on the optimal offloading decision; The optimal offloading decision determination module includes: a terminal-base station relationship judgment submodule, configured to judge whether the future stay area of ​​the terminal is in the coverage overlap area of ​​the base station within a preset range from the future stay area; a first task offloading submodule, configured to obtain a first optimal offloading decision based on the distance information between the future stay area of ​​the terminal and the base station within the preset range from the future stay area through a first optimal energy consumption model if the future stay area of ​​the terminal is not in the coverage overlap area of ​​the base station within the preset range from the future stay area, and to unload the task to be executed from the terminal according to the first optimal offloading decision; a second task offloading submodule, configured to obtain a second optimal offloading decision based on the distance information between the future stay area of ​​the terminal and the base station within the preset range from the future stay area and base station influencing factors through a second optimal energy consumption model if the future stay area of ​​the terminal is in the coverage overlap area of ​​the base station within the preset range from the future stay area, and to unload the task to be executed from the terminal according to the second optimal offloading decision; The second task offloading submodule includes: a distance information and base station influencing factor input submodule, which is used to input the distance information between the future stay area of ​​the terminal and the base station within the preset range of the future stay area and the base station influencing factor into the second optimal energy consumption model; an optimized simulated annealing intelligent algorithm calculation submodule, which is used to obtain the second optimal offloading decision using the optimized simulated annealing intelligent algorithm through the second optimal energy consumption model; and a second offloading submodule, which is used to offload the to-be-executed task from the terminal to the corresponding base station according to the second optimal offloading decision. The information contained in the second optimal energy consumption model is: Wherein, m represents the number of base stations whose distances between the future stay area of ​​the terminal and the base stations within the preset range from the future stay area are less than a preset threshold, s i represents the utility score of base station i; q represents the number of base station influencing factors that affect the transmission rate, including the base station's received signal strength, the amount of task data processed by the base station, and the base station's maximum allowable delay; Limit j (j=1, 2, ..., q) represents the index constraint corresponding to each of the base station influencing factors; c qi represents the energy consumption of the i-th base station under the influence of the q-th base station, x i Indicates whether base station i is available.

7. An electronic device comprising a processor and a memory storing a computer program, characterized in that: When the processor executes the computer program, the steps of the task offloading method according to any one of claims 1 to 5 are implemented.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the task offloading method according to any one of claims 1 to 5 are implemented.

Images