Cellular network task unloading method based on mobile edge computing

Abstract

The invention provides a cellular network task unloading method based on mobile edge computing. The method comprises the following steps: constructing a mobile edge computing unloading system architecture model; establishing a communication model from the mobile terminal to the base station according to the Shannon-Hartley theorem; respectively establishing a corresponding time delay calculation model and an energy consumption calculation model according to the task execution mode of the mobile terminal; establishing a base station selection strategy in a multi-terminal and multi-server scene; and calculating a subtask unloading strategy of the mobile terminal by adopting an improved artificial fish swarm algorithm, so that each subtask finds an optimal unloading strategy. According to the method, the problems of high time delay and low efficiency caused by long distance between the mobile terminal and the cloud server are effectively solved, and efficient and accurate task unloading is realized through the optimal unloading strategy.

Description

Technical Field

[0001] This invention relates to the field of mobile communications, and more specifically to a cellular network task offloading method based on mobile edge computing. Background Technology

[0002] With the booming development of smart devices such as online games, augmented reality, the Internet of Things, and connected vehicles, a large number of computing tasks have been generated on resource-constrained devices. Compute offloading technology provides a solution for this. A cellular network is a mobile communication hardware architecture. Because the signal coverage of the communication base stations constituting the network coverage is hexagonal, the entire network resembles a honeycomb, hence its name. In cellular networks, compute offloading technology was initially applied to Mobile Cloud Computing (MCC). The advantage of MCC is that it provides ample computing resources, alleviating the high energy consumption and high latency of executing tasks locally.

[0003] However, cloud computing also has significant drawbacks. Because mobile devices are geographically distant from cloud servers, large transmission distances can cause significant latency when performing heavy tasks, leading to increased energy consumption. This makes it unsuitable for latency-sensitive applications. To address the high latency issue caused by cloud computing, Mobile Edge Computing (MEC) was proposed. This method offloads tasks from mobile devices to a server located closer to the MEC via a wireless cellular network, significantly improving both computing speed and quality.

[0004] Computational offloading is a key technology in mobile edge computing. However, existing technologies rarely combine the characteristics of MCC and MEC for task offloading in multi-terminal device and multi-server scenarios. Furthermore, while the algorithms converge quickly in the early stages, their convergence speed slows significantly in later stages, easily leading to local optima and low accuracy, thus hindering application effectiveness. Therefore, there is an urgent need to invent a more efficient and accurate task offloading method for cellular networks. Summary of the Invention

[0005] The purpose of this invention is to provide a cellular network task offloading method based on mobile edge computing, comprising:

[0006] A mobile edge computing offloading system architecture model is constructed, which includes a remote cloud server, multiple base stations configured with edge servers, and multiple mobile terminals;

[0007] A communication model between the mobile terminal and the base station is established based on the Shannon-Hartley theorem.

[0008] Establish corresponding latency calculation models and energy consumption calculation models based on the task execution methods of mobile terminals;

[0009] Establish a base station selection strategy for multi-terminal and multi-server scenarios to enable optimal base station selection when the mobile terminal is covered by multiple base stations.

[0010] An improved artificial fish swarm algorithm is used to calculate the subtask unloading strategy of the mobile terminal so that each subtask finds the optimal unloading strategy.

[0011] As an example, the task execution methods of mobile terminals include local execution, edge server execution, and remote cloud server execution.

[0012] As an example, a task can be broken down into multiple sub-tasks, and the execution method of each sub-task can be decided and deployed independently.

[0013] Exemplary examples include local computing models, edge computing models, and cloud computing models;

[0014] The local execution unloading delay is the computation processing time of the mobile terminal's CPU;

[0015] The offloading latency performed by the edge server includes edge computing transmission latency and edge computing processing latency. The edge computing transmission latency includes the sum of the latency of the subtask from the mobile terminal to the edge server and the latency of the execution result being fed back from the edge server to the mobile terminal.

[0016] The offloading latency of cloud computing includes cloud computing transmission latency and cloud computing processing latency. The cloud computing transmission latency includes the sum of the transmission latency of the subtask from the mobile terminal to the edge server and then from the edge server to the remote cloud server, and the transmission latency of the execution result from the remote cloud server to the edge server and then from the edge server to the mobile terminal.

[0017] As an example, the energy consumption calculation model is the product of task execution delay and circuit power.

[0018] Exemplary, the establishment of a base station selection strategy in a multi-terminal, multi-server scenario, to enable optimal base station selection when the mobile terminal is covered by multiple base stations, includes:

[0019] Obtain the base station load rate, distance between the terminal device and the base station, base station signal coverage, and overhead data of subtasks offloading to the base station;

[0020] Based on the base station load rate, the distance between the terminal equipment and the base station, the base station signal coverage, and the overhead data of subtasks offloading to the base station, a base station selection function is established.

[0021] Determine whether the mobile terminal is covered by multiple base stations;

[0022] If so, the optimal base station is determined according to the base station selection function.

[0023] Exemplary, the subtask offloading strategy for the mobile terminal calculated using the improved artificial fish swarm algorithm includes:

[0024] The task to be processed is divided into multiple subtasks, wherein each subtask is a fully functional task to be uninstalled.

[0025] The improved artificial fish swarm algorithm is used to select the unloading decision for each subtask;

[0026] Based on the selected unloading decision, subtasks to be executed locally are set to execute locally, subtasks to be executed at the edge are uploaded to the edge server for execution, and subtasks to be executed in the cloud are uploaded to the remote cloud server for execution. After execution is completed, the calculation results are transmitted back to the terminal device.

[0027] Exemplary, the selection of unloading decisions for each subtask using the improved artificial fish swarm algorithm includes:

[0028] The improved artificial fish swarm algorithm is obtained by optimizing the artificial fish swarm algorithm using the wolf pack algorithm.

[0029] This invention proposes a cellular network task offloading method based on mobile edge computing, including constructing a mobile edge computing offloading system architecture model; establishing a communication model from the mobile terminal to the base station based on the Shannon-Hartley theorem; establishing corresponding latency and energy consumption calculation models according to the task execution mode of the mobile terminal; establishing a base station selection strategy in a multi-terminal, multi-server scenario; and using an improved artificial fish swarm algorithm to calculate the sub-task offloading strategy of the mobile terminal, so that each sub-task finds an optimal offloading strategy. This invention effectively solves the problems of high latency and low efficiency caused by the long distance between the mobile terminal and the cloud server, and achieves efficient and accurate task offloading through the optimal offloading strategy. Attached Figure Description

[0030] Figure 1 This is a flowchart of a cellular network task offloading method based on mobile edge computing provided in an embodiment of the present invention;

[0031] Figure 2 A flowchart of another cellular network task offloading method based on mobile edge computing provided in an embodiment of the present invention. Detailed Implementation

[0032] The technical solution of the present invention will be clearly and completely described below with reference to specific embodiments. However, those skilled in the art should understand that the embodiments described below are only for illustrating the present invention and should not be regarded as limiting the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] The tasks mentioned in the embodiments of this invention can be decomposed into multiple sub-tasks, and the execution method of each sub-task is decided and deployed independently. (Refer to...) Figure 1 and Figure 2 The method of this invention includes:

[0034] Step S101: Construct a mobile edge computing offloading system architecture model. The mobile edge computing offloading system architecture model includes a remote cloud server, multiple base stations configured with edge servers, and multiple mobile terminals.

[0035] Specifically, the system architecture of this embodiment consists of a remote cloud server, multiple base stations configured with edge servers, and multiple mobile terminal devices, forming a three-layer MEC computing offloading structure of "cloud-edge-device". The terminal devices are connected to the base station via a wireless link, and the base station is connected to the cloud via a wired link. Mobile terminal tasks can be executed locally or offloaded to the edge of the base station for execution. The base station can offload tasks to the remote cloud server, receive the computing results from the remote cloud server, and then return them to the terminal devices.

[0036] Step S102: Establish a communication model from the mobile terminal to the base station based on the Shannon-Hartley theorem.

[0037] Specifically, this embodiment of the invention employs a code division multiple access (CDMA) channel access scheme, using different code sequences to achieve communication. The terminal device transmits data to the base station via a wireless channel. According to the Shannon-Hartley theorem, the communication model from the terminal device to the base station is defined as follows:

[0038] (1)

[0039] Where B is the channel bandwidth. It is background noise inside the channel. It is the transmission power of terminal device i. It is channel gain. This interference is caused by other terminal devices to terminal device i. Because the edge server is located on the base station, the data transmission rate from the terminal device to the base station is the same as the transmission rate to the edge server.

[0040] Step S103: Establish corresponding latency calculation models and energy consumption calculation models according to the task execution methods of the mobile terminal. Here, the task execution methods of the mobile terminal include local execution, edge server execution, and remote cloud server execution.

[0041] Specifically, the latency calculation model includes a local computing model, an edge computing model, and a cloud computing model. The offload latency for local execution is the computational processing time of the mobile terminal's CPU; the offload latency for edge server execution includes edge computing transmission latency and edge computing processing latency, wherein the edge computing transmission latency includes the sum of the latency of the subtask from the mobile terminal to the edge server and the latency of the execution result being fed back from the edge server to the mobile terminal; the offload latency for cloud computing includes cloud computing transmission latency and cloud computing processing latency, wherein the cloud computing transmission latency includes the sum of the transmission latency of the subtask from the mobile terminal to the edge server, and then from the edge server to the remote cloud server, and the transmission latency of the execution result being fed back from the remote cloud server to the edge server, and then from the edge server to the mobile terminal. The energy consumption calculation model is the product of task execution latency and circuit power.

[0042] Step S104: Establish a base station selection strategy for multi-terminal and multi-server scenarios to enable optimal base station selection when the mobile terminal is covered by multiple base stations.

[0043] Specifically, this step mainly includes obtaining base station load rate, distance between terminal device and base station, base station signal coverage area, and overhead data of subtasks offloading to base station; establishing a base station selection function based on the base station load rate, distance between terminal device and base station, base station signal coverage area, and overhead data of subtasks offloading to base station; determining whether the mobile terminal is covered by multiple base stations; if so, determining the optimal base station based on the base station selection function.

[0044] For multi-terminal device-multi-server scenarios, when a terminal device is covered by multiple base stations, it is necessary to select one of the base stations for subtask offloading. A base station selection strategy is proposed, as shown in formula (2).

[0045] The base station selection function is determined by the base station load rate k and the distance between the terminal device and the base station. Base station signal coverage area and the overhead of offloading subtasks to the edge. Decision. When a base station is near full load, the load rate is approximately 1, and the base station selection function is approximately 0. In this case, it is not advisable to offload the base station.

[0046] (2)

[0047] Step S105: The improved artificial fish swarm algorithm is used to calculate the subtask unloading strategy of the mobile terminal so that each subtask finds the optimal unloading strategy.

[0048] Specifically, the artificial fish swarm algorithm converges quickly in the early stages but slowly in the later stages, easily leading to local optima and low accuracy. Therefore, this embodiment of the invention optimizes the artificial fish swarm algorithm using a wolf pack algorithm to select better individuals. The wolf pack algorithm simulates the hunting behavior of a wolf pack to address the optimization problem of the objective function. Each artificial wolf is considered a feasible solution to the problem, and the prey odor concentration is considered the fitness value of the objective function. The wolf pack is divided into alpha wolves, scout wolves, and predator wolves. In this embodiment, the artificial fish are also divided into alpha fish, scout fish, and predator fish, whose roles are the same as in the wolf pack algorithm. After the artificial fish swarm engages in foraging, grouping, tail chasing, and random behavior, the wolf pack optimization behavior is performed, using the wolf pack algorithm to select better individuals within the swarm.

[0049] This step mainly includes dividing the task to be processed into multiple subtasks, where each subtask is a fully functional task to be uninstalled; using an improved artificial fish swarm algorithm to select an uninstallation decision for each subtask; based on the selected uninstallation decision, setting the subtasks to be executed locally, uploading the subtasks to be executed at the edge to the edge server, uploading the subtasks to be executed in the cloud to the remote cloud server, and transmitting the calculation results back to the terminal device after execution.

[0050] This invention discloses a cellular network task offloading method based on mobile edge computing, including constructing a mobile edge computing offloading system architecture model; establishing a communication model from the mobile terminal to the base station based on the Shannon-Hartley theorem; establishing corresponding latency calculation models and energy consumption calculation models according to the task execution mode of the mobile terminal; establishing a base station selection strategy in a multi-terminal, multi-server scenario; and using an improved artificial fish swarm algorithm to calculate the sub-task offloading strategy of the mobile terminal, so that each sub-task finds an optimal offloading strategy. This invention effectively solves the problems of high latency and low efficiency caused by the long distance between the mobile terminal and the cloud server, and achieves efficient and accurate task offloading through the optimal offloading strategy.

[0051] Although the present invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A method for offloading cellular network tasks based on mobile edge computing, characterized in that, include: A mobile edge computing offloading system architecture model is constructed, which includes a remote cloud server, multiple base stations configured with edge servers, and multiple mobile terminals; A communication model between the mobile terminal and the base station is established based on the Shannon-Hartley theorem. Establish corresponding latency calculation models and energy consumption calculation models based on the task execution methods of mobile terminals; Establish a base station selection strategy for multi-terminal and multi-server scenarios to enable optimal base station selection when the mobile terminal is covered by multiple base stations. An improved artificial fish swarm algorithm is used to calculate the subtask unloading strategy of the mobile terminal so that each subtask finds the optimal unloading strategy.

2. The cellular network task offloading method based on mobile edge computing according to claim 1, characterized in that, The task execution methods of the mobile terminal include local execution, edge server execution, and remote cloud server execution.

3. The cellular network task offloading method based on mobile edge computing according to claim 2, characterized in that, The task can be broken down into multiple subtasks, and the execution method of each subtask is decided and deployed independently.

4. The cellular network task offloading method based on mobile edge computing according to claim 3, characterized in that, The latency calculation model includes a local computing model, an edge computing model, and a cloud computing model; The local execution unloading delay is the computation processing time of the mobile terminal's CPU; The offloading latency performed by the edge server includes edge computing transmission latency and edge computing processing latency. The edge computing transmission latency includes the sum of the latency of the subtask from the mobile terminal to the edge server and the latency of the execution result being fed back from the edge server to the mobile terminal. The unloading latency of the cloud server includes cloud computing transmission latency and cloud computing processing latency. The cloud computing transmission latency includes the sum of the transmission latency of the subtask from the mobile terminal to the edge server and then from the edge server to the remote cloud server, and the transmission latency of the execution result from the remote cloud server to the edge server and then from the edge server to the mobile terminal.

5. The cellular network task offloading method based on mobile edge computing according to claim 4, characterized in that, The energy consumption calculation model is the product of task execution delay and circuit power.

6. The cellular network task offloading method based on mobile edge computing according to claim 5, characterized in that, The establishment of a base station selection strategy in a multi-terminal, multi-server scenario, to enable optimal base station selection when the mobile terminal is covered by multiple base stations, includes: Acquire base station load rate, distance between terminal equipment and base station, base station signal coverage, and overhead data of subtasks offloaded to base station, and establish a base station selection function; Determine whether the mobile terminal is covered by multiple base stations; If so, the optimal base station is determined according to the base station selection function.

7. The cellular network task offloading method based on mobile edge computing according to claim 6, characterized in that, The subtask unloading strategy for the mobile terminal, calculated using the improved artificial fish swarm algorithm, includes: The task to be processed is divided into multiple subtasks, wherein each subtask is a fully functional task to be uninstalled. The improved artificial fish swarm algorithm is used to select the unloading decision for each subtask; Based on the selected unloading decision, subtasks to be executed locally are set to execute locally, subtasks to be executed at the edge are uploaded to the edge server for execution, and subtasks to be executed in the cloud are uploaded to the remote cloud server for execution. After execution is completed, the calculation results are transmitted back to the terminal device.

8. The cellular network task offloading method based on mobile edge computing according to claim 7, characterized in that, The selection of unloading decisions for each subtask using the improved artificial fish swarm algorithm includes: The improved artificial fish swarm algorithm is obtained by optimizing the artificial fish swarm algorithm using the wolf pack algorithm.

Images