A method for task unloading and resource allocation in D2D-assisted MEC

Abstract

This invention discloses a task offloading and resource allocation method under D2D-assisted MEC, belonging to the field of wireless communication and computing resource management technology. The method introduces service-oriented devices assisted by terminal direct transmission communication technology and rationally selects task processing modes, achieving efficient resource utilization and avoiding resource waste. By jointly optimizing computation offloading, D2D selection, computing resource allocation, and spectrum resource allocation, it can both guarantee service quality and reduce network congestion, thereby improving user experience. The two-stage iterative algorithm of this invention combines block coordinate descent, reconstruction linearization technology, and convex optimization methods, effectively solving non-convex optimization problems, and has a fast convergence speed and low computational complexity. Furthermore, the D2D-assisted MEC architecture and optimization method are applicable to different types of intelligent devices and diverse application scenarios, possessing broad applicability and flexibility, and can meet the needs of future mobile edge computing development.

Description

Technical Field

[0001] This invention relates to a method for task offloading and resource allocation in D2D assisted MEC, belonging to the field of wireless communication and computing resource management technology. Background Technology

[0002] With the proliferation of smart devices (SDs) in modern communication and computing, the demand for computing power and storage capacity is growing daily. However, limited by hardware conditions, a single smart device struggles to efficiently handle complex computing tasks. To address this issue, mobile edge computing (MEC) has become a key technology, helping smart devices process computing tasks by deploying computing resources at the network edge.

[0003] The introduction of Device-to-Device (D2D) communication technology enables MEC architecture to communicate directly between devices and share computing resources, further improving resource utilization efficiency. D2D communication not only provides lower latency communication services, but also reduces the computing burden on intelligent devices by enabling them to collaborate on computing tasks, thereby enhancing the overall system's computing power and resource utilization efficiency.

[0004] In traditional MEC systems, resource management schemes typically rely on a single compute offloading mode, failing to fully utilize various network resources. This results in low resource utilization, severe network congestion, and compromised Quality of Service (QoS). Therefore, designing a scheme that can fully utilize multiple resources and optimize offloading strategies has become an urgent problem to be solved. Summary of the Invention

[0005] To address the issues of low resource utilization and high execution costs in current mobile edge computing (MEC) systems, this invention provides a method for task offloading and resource allocation in D2D-assisted MEC, the technical solution of which is as follows:

[0006] The task unloading and resource allocation method of the present invention includes:

[0007] Step 1: Construct a D2D communication-assisted MEC system model, define the task processing mode of intelligent devices and the computing task offloading mode of edge servers, and divide the tasks generated by task-oriented intelligent devices into three task processing modes: local computing, D2D offloading and edge offloading.

[0008] Step 2: With minimizing task execution cost as the optimization objective, the task partitioning strategy, computing resource allocation strategy, terminal direct transmission offloading selection strategy, and dynamic pricing strategy are jointly formulated.

[0009] Step 3: The joint optimization problem is decomposed into multiple subproblems using the block coordinate descent method and the reconstruction-linearization technique. The non-convex problem is approximated as a convex optimization problem, and the convex optimization theory is applied to further optimize the task unloading ratio and the computing resource allocation strategy.

[0010] Optionally, the task execution cost in step 2 is:

[0011]

[0012] in, m represents the number of computation bits allocated to tasks of local computation, D2D offloading, and edge offloading, respectively. l , and m e These represent the unit price for providing computing services per unit bit by task-oriented intelligent devices, service-oriented intelligent devices, and edge servers, respectively. i,j ∈{0,1} is a binary choice variable, x i,j =1 indicates a task-oriented intelligent device With service-oriented smart device SSDs There is a D2D communication link, x i,j =0 indicates that there is no D2D communication link between the two;

[0013] The task partitioning strategy, computing resource allocation strategy, terminal direct download offloading selection strategy, and dynamic pricing strategy are jointly formulated as follows:

[0014]

[0015] Among them, L i This refers to task-oriented smart devices (TSDs). The number of computation bits generated by the task. SSDs The computation frequency of executing local tasks. SSDs Assigned to TSDs The calculation frequency, SSDs The maximum frequency; This indicates that BS is allocated to SSDs. The calculation frequency, Indicates the maximum frequency of BS; T represents the time delay between the two phases of D2D unloading and edge unloading. i This indicates the maximum tolerable time threshold. This indicates that TSDs have undergone three stages of processing: local computation, D2D unloading, and edge unloading. The remaining battery energy.

[0016] Optionally, step 3 includes:

[0017] Step 31: Given the selection strategy X m Solve the following problem to obtain the locally optimal unloading scheme L for the current iteration number. m +1 and frequency calculation scheme F m+1 :

[0018]

[0019] Where, μ i,j As an auxiliary variable, C i Indicates TSDs CPU cycle count required to process a unit bit task; C j SSDs The number of CPU cycles required to process a unit bit task; Indicates the data transfer rate during the D2D offloading phase; L j Indicates the total data size of the tasks pending execution on the SSDs; ν i Represents auxiliary variables. Indicates the data transfer rate during edge offloading; TSDs represent the D2D unloading phase. The transmission power; Indicates edge unloading phase TSDs Transmit power when transmitting data to the BS; E i Indicates TSDs Initial energy state; This indicates the energy consumed during the local computation phase;

[0020] Step 32: Based on the locally optimal unloading scheme L obtained in step 31 m+1 and frequency calculation scheme F m+1 Calculate the locally optimal choice strategy X m+1 The solution is as follows:

[0021]

[0022] The above equation is solved using a solver to obtain a feasible solution;

[0023] Step 33: Iterate through steps 31 and 32 until an unloading scheme, calculation frequency, and selection strategy that meet the preset conditions are obtained.

[0024] Optionally, the communication process employs orthogonal frequency division multiple access (OFDMA) technology.

[0025] Optionally, step 3 uses the CVX solver to solve the convex optimization problem.

[0026] Optionally, the data transmission rate during the D2D offloading phase... The calculation method is as follows:

[0027]

[0028] Where B represents the transmission bandwidth of each sub-channel. and They represent TSDs respectively The transmit power and D2D link channel power gain, where N represents the noise power of each sub-channel.

[0029] Optional, data transfer rate during edge offloading The calculation formula is:

[0030]

[0031] Where B represents the transmission bandwidth of each sub-channel. and They represent TSDs respectively Transmit power and cellular link channel gain when transmitting data to SSDs via D2D link, where N represents the noise power of each sub-channel.

[0032] This invention provides a task unloading and resource allocation system, the system being used to implement the method described in any of the preceding claims, the system comprising:

[0033] The system model building module is configured to build a D2D communication-assisted MEC system model, define the task processing mode of intelligent devices and the computing task offloading mode of edge servers, and divide the tasks generated by task-oriented intelligent devices into three task processing modes: local computing, D2D offloading and edge offloading.

[0034] The joint optimization formula construction module is configured to jointly formulate the task partitioning strategy, computing resource allocation strategy, terminal direct transmission offloading selection strategy and dynamic pricing strategy with the optimization objective of minimizing task execution cost.

[0035] The solution module is configured to decompose the joint optimization problem into multiple subproblems using the block coordinate descent method and the reconstruction-linearization technique, approximate the non-convex problem as a convex optimization problem, and apply convex optimization theory to further optimize the task offloading ratio and computational resource allocation strategy.

[0036] This invention provides an electronic device, characterized in that it includes a memory and a processor;

[0037] The memory is used to store computer programs;

[0038] The processor is configured to, when executing the computer program, implement the method as described in any of the preceding methods.

[0039] The present invention provides a computer-readable storage medium, characterized in that the storage medium stores a computer program, which, when executed by a processor, implements the method described in any of the preceding claims.

[0040] The beneficial effects of this invention are:

[0041] This invention optimizes computation offloading and resource allocation strategies to minimize user overhead in communication and computing services, thereby effectively reducing the overall execution cost for edge users. By introducing terminal-to-device (D2D) communication technology-assisted service devices (SSDs) and rationally selecting task processing modes (local execution, D2D offloading, edge computing), efficient resource utilization is achieved, avoiding resource waste. This invention jointly optimizes computation offloading, D2D selection, computing resource allocation, and spectrum resource allocation, ensuring both Quality of Service (QoS) and reducing network congestion, thus improving user experience. The designed two-stage iterative algorithm combines block coordinate descent, Reconstruction Linearization (RLT), and convex optimization methods, effectively solving non-convex optimization problems with fast convergence speed and low computational complexity. Furthermore, the D2D-assisted MEC architecture and optimization methods are applicable to different types of smart devices and diverse application scenarios, exhibiting broad applicability and flexibility, and meeting the future needs of mobile edge computing development. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a graph showing the relationship between the number of TSDs and the total system cost under different unloading schemes according to the present invention.

[0044] Figure 2 This is a graph showing the relationship between the maximum tolerable latency and the total system cost under different unloading schemes.

[0045] Figure 3 This is a graph showing the relationship between task data size and total system cost under different unloading schemes according to the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0047] Example 1:

[0048] This embodiment provides a method for task unloading and resource allocation, including:

[0049] Step 1: Construct a D2D communication-assisted MEC system model, define the task processing mode of intelligent devices and the computing task offloading mode of edge servers, and divide the tasks generated by task-oriented intelligent devices into three task processing modes: local computing, D2D offloading and edge offloading.

[0050] Step 2: With minimizing task execution cost as the optimization objective, the task partitioning strategy, computing resource allocation strategy, terminal direct transmission offloading selection strategy, and dynamic pricing strategy are jointly formulated.

[0051] Step 3: The joint optimization problem is decomposed into multiple subproblems using the block coordinate descent method and the reconstruction-linearization technique. The non-convex problem is approximated as a convex optimization problem, and the convex optimization theory is applied to further optimize the task unloading ratio and the computing resource allocation strategy.

[0052] Example 2:

[0053] This embodiment provides a method for offloading multiple tasks in a MEC environment based on terminal direct transmission communication assistance. The applicable system model is an MEC-assisted D2D communication system consisting of a base station (BS) and several smart devices (SDs), with an edge server deployed on the BS. Based on whether local computing resources can complete tasks on time, all SDs in the system are divided into two categories: task-oriented smart devices (TSDs) and service-oriented smart devices (SSDs), denoted as follows: and SSDs can establish independent D2D links with TSDs to assist TSDs. The task must be completed within the maximum tolerable time threshold. If some tasks cannot be computed locally on TSDs, they can be offloaded to edge servers via cellular links or to SSDs via D2D links.

[0054] Suppose TSDs generate a data tuple (C) for a computationally intensive, latency-critical task. i ,L i ,T i ), Where C i L represents the number of CPU cycles required to process a unit of bit task. i T represents the total amount of data for the current task.i This indicates the maximum tolerable latency limit for task execution. Furthermore, to incentivize SSDs to serve more users, TSDs employ a dynamic pricing mechanism to charge for computing resources used during the offloading process, alleviating congestion issues caused by limited resources in a multi-user environment. Tasks on TSDs are divided into three parts: some tasks are computed locally, some tasks are offloaded to SSDs for computation, and some tasks are offloaded to edge servers for computation. and These represent the number of computation bits allocated to the three stages: local computation, D2D offloading, and edge offloading. Clearly, different task allocation ratios affect the battery energy and resource costs of TSDs. Generate L i The task, and in The task is performed locally. The task is executed on SSDs via a D2D link, while The task is unloaded to the BS process.

[0055] The computational unloading and resource allocation method in this embodiment includes:

[0056] Step 1: Construct a MEC system model assisted by terminal direct transmission communication (D2D), and define the task processing mode of smart devices and the computing task offloading mode of edge servers.

[0057] Using decision matrix Describe the D2D link relationship, where x i,j ∈{0,1} is a binary choice variable, x i,j =1 indicates TSDs With SSDs A D2D communication link was successfully established, and x i,j =0 indicates that there is no D2D communication link between the two. and These represent the number of computation bits allocated to the three stages: local computation, D2D offloading, and edge offloading, respectively. TSDs Generate L i The task, and in The task is performed locally. The task is executed on SSDs via a D2D link, while The task is unloaded to the BS for processing. (For TSDs) In terms of latency, the three stages of local computation, D2D offloading, and edge offloading are adopted. This indicates that the latency in both the D2D offloading and edge offloading phases also includes two states: transmission and computation, denoted as... In time block Ti Within this framework, assuming that parameters such as channel allocation and transmit power are constant, the transmission bandwidth and noise power of each sub-channel are labeled as B and N, respectively.

[0058] 1) D2D Offload Communication: Utilizing SSDs' computing resources, offloaded data packets are transferred from TSDs. Direct transfer to SSDs The above is executed. During this process, the transmit power of TSDs and the power gain of the D2D link channel paired with (i,j) are defined as... and Calculate the data transfer rate during the D2D offloading phase using Shannon's theorem. Right now:

[0059]

[0060] The time and energy consumption of the D2D offloading phase are mainly affected by the D2D link transmission. The impact of TSDs tasks. The latency and energy consumption expressions for this process are as follows:

[0061]

[0062] 2) Edge offloading communication: SSDs If local computing cannot complete the task in time, it will be offloaded to the BS via the cellular link. This is similar to the D2D offloading phase definition. and Represented as TSDs Transmit power and cellular link channel gain when transmitting data to the BS. Therefore, data transmission rate during edge offloading. The calculation is as follows:

[0063]

[0064] Only perform edge offloading processing on data packets Therefore, TSDs are calculated. The mathematical expressions for transmission delay and energy loss are as follows:

[0065]

[0066] Because BS and SSDs have certain computing and storage resources, TSDs It provides limited computing, caching, and storage services, and rationally allocates resources to maximize utilization. It assumes that both TSDs and SSDs use simplex communication mode, performing data processing upon successful data reception. Based on different computing modes, the computing model can be divided into two parts: a local computing model and an offloaded model.

[0067] 1) Local computing

[0068] Under conditions of limited energy consumption and resources, TSDs will fully utilize their own computing resources to reduce the proportion of computing and communication resources consumed by offloading data. The time consumed in the local computing phase is:

[0069]

[0070] Among them, f i For TSDs The CPU frequency. To maximize the use of local resources, assume... The local computing task data size is The energy consumption during the local computing phase is:

[0071]

[0072] Among them κ i This represents the effective switched capacitor coefficient.

[0073] 2) D2D unloading calculation

[0074] SSDs Received uninstallation After receiving the data packet, the SSDs utilize their remaining computing resources to process the packet and return the results to the TSDs. The time spent by the SSDs executing tasks during the D2D unloading process... Represented as:

[0075]

[0076] in, For SSDs The computation frequency invoked. It's worth noting that SSDs may generate pending tasks, thus requiring local tasks to be executed first before assisting TSDs in unloading. Therefore, a prerequisite for D2D unloading is that the SSDs complete their local tasks and receive the unloading data packet. Using... Indicating the local computation time of SSDs, the total latency of the D2D unloading phase is:

[0077]

[0078] in,

[0079] 3) Edge unloading calculation

[0080] TSDs successfully transmitted to BS After the data packet, the edge server uses SSDs. Allocate computing resources appropriately and assist in unloading the task. The time consumed in this process is:

[0081]

[0082] in, Allocate SSDs to BS Calculate the frequency. At this point, the total edge unloading delay is:

[0083]

[0084] Step 2: Establish a joint optimization problem for multi-task partial offloading, and jointly formulate the task partitioning strategy, computing resource allocation strategy, terminal direct transmission offloading selection strategy, and dynamic pricing strategy.

[0085] For local computing tasks D2D Unloading Task and edge unloading tasks Define TSDs SSDs The unit price for BS to provide computing services per unit bit is m. l , and m e To complete the task within the specified time, TSDs within the D2D-MEC system Execution cost M i Represented as:

[0086]

[0087] Assuming TSDs The initial energy state is labeled E i After three stages of processing—local computation, D2D unloading, and edge unloading—TSDs Remaining battery energy status Calculated as

[0088]

[0089] By jointly optimizing task offloading schemes, computing resource allocation, and D2D association strategies, the execution cost problem is modeled as follows:

[0090]

[0091] Where constraints C1 and C2 represent the limitations satisfied by the task partitioning strategy, C3, C4, and C5 restrict the execution of tasks within a specified time and CPU frequency range, C6 indicates the battery energy state that TSDs need to maintain, and C7 and C8 represent the D2D association strategy. Clearly, due to the discrete optimization variable x... i,j Since the objective function and constraints C7 and C9 are non-convex, it can be determined that the objective function P1 is a non-convex problem, which is difficult to solve directly using traditional convex optimization techniques.

[0092] Step 3: The joint optimization problem is decomposed into multiple subproblems using the block coordinate descent method and reconstruction-linearization (RLT) technique. The non-convex problem is approximated as a convex optimization problem, and convex optimization theory is applied to further optimize the task offloading ratio and computational resource allocation strategy.

[0093] use This represents the selection strategy for all SSDs, given an initial strategy X. 0 For distance-based selection strategies, X l (l = 1, 2, ..., L) represents the strategy chosen in the l-th iteration. Given a strategy X... l At that time, SSDs Resources supplied to TSDs This establishes a pairing relationship (i,j) between devices, at which point a D2D association matrix x exists. i,j =1. Combining formulas (10), (12), and (14), P1 can be rewritten as:

[0094]

[0095] Because there is variable coupling in C5 and C7 and Therefore, problem P2 is nonconvex, and the product terms can be transformed using the reformulation-linearization technique (RLT). An auxiliary variable μ is introduced. i,j and ν i ,make but

[0096] μ i,j Substituting the RLT constraint factor product limit Simplifying, we get:

[0097]

[0098] Similarly, substitute We can then conclude that:

[0099]

[0100] Given a selection strategy X m Solve the following problem to obtain the locally optimal unloading scheme L for the current iteration number. m+1 and frequency calculation scheme F m+1 :

[0101]

[0102] Analysis shows that the transformed objective function P2.2 is a convex optimization problem, which can be solved using convex optimization techniques, such as the CVX toolbox, to obtain the local optimum solution for the current iteration number.

[0103] After resource allocation, a locally optimal solution is obtained. and Based on this, the original optimization problem P1 is transformed into a SSDs selection subproblem. For the discrete variable x... i,j The relaxed constraint is 0 ≤ x i,j ≤1. Based on the locally optimal unloading scheme L m+1 and frequency calculation scheme F m+1 Calculate the locally optimal choice strategy X m+1 The solution is as follows:

[0104]

[0105] After transformation, problem P3.1 is convex, and a feasible solution can be obtained using the CVX solver. The optimization variable x is then optimized. i,j The specific iterative steps for optimizing the system execution cost during recovery are shown in Table 1.

[0106] Table 1 Task Unloading and Resource Allocation Algorithm

[0107]

[0108]

[0109] To make the objectives, technical solutions, and advantages of this invention clearer, the following comparison will be made between some classic detection algorithms and the method proposed in this invention, demonstrating the performance superiority of the computation offloading and resource allocation method under the D2D-assisted MEC architecture of this invention.

[0110] The baseline algorithms used for simulation were the traditional partial offloading scheme, the cooperative computing offloading scheme, and the relay-assisted offloading scheme. Experimental results are as follows: Figure 1 , 2 As shown in Figure 3, the simulation curves show that by comparing the total cost under different numbers of smart devices, maximum tolerable latency, and task data size, the computational offloading and resource allocation method of the present invention can effectively optimize the total system cost.

[0111] In summary, simulation results verify the performance of the joint optimization method of this invention in different scenarios, and compared with the baseline algorithm, this algorithm can significantly reduce execution costs. Therefore, the computational offloading and resource allocation method of this invention can effectively improve network system performance while completing low-latency tasks.

[0112] Some steps in the embodiments of the present invention can be implemented using software, and the corresponding software program can be stored in a readable storage medium, such as an optical disc or a hard disk.

[0113] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for task unloading and resource allocation, characterized in that, The method includes: Step 1: Construct a D2D communication-assisted MEC system model, define the task processing mode of intelligent devices and the computing task offloading mode of edge servers, and divide the tasks generated by task-oriented intelligent devices into three task processing modes: local computing, D2D offloading and edge offloading. Step 2: With minimizing task execution cost as the optimization objective, the task partitioning strategy, computing resource allocation strategy, terminal direct transmission offloading selection strategy, and dynamic pricing strategy are jointly formulated. Step 3: The joint optimization problem is decomposed into multiple sub-problems using the block coordinate descent method and the reconstruction-linearization technique. The non-convex problem is approximated as a convex optimization problem, and the convex optimization theory is applied to further optimize the task offloading ratio and the computing resource allocation strategy. The task execution cost in step 2 is: in, m represents the number of computation bits allocated to tasks of local computation, D2D offloading, and edge offloading, respectively. l , and m e These represent the unit price for providing computing services per unit bit by task-oriented intelligent devices, service-oriented intelligent devices, and edge servers, respectively. i,j ∈{0,1} is a binary choice variable, x i,j =1 indicates a task-oriented intelligent device With service-oriented smart devices There is a D2D communication link, x i,j =0 indicates that there is no D2D communication link between the two; The task partitioning strategy, computing resource allocation strategy, terminal direct download offloading selection strategy, and dynamic pricing strategy are jointly formulated as follows: Among them, L i Indicates task-oriented intelligent devices The number of computation bits generated by the task. SSDs The computation frequency of executing local tasks. express Assigned to The calculation frequency, express Maximum frequency; f i e Indicates that BS is assigned to The calculation frequency, Indicates the maximum frequency of BS; T i k ,k={d,e} represents the time delay of the two stages of D2D unloading and edge unloading, T i This indicates the maximum tolerable time threshold. This indicates that after processing through three stages: local computation, D2D unloading, and edge unloading. The remaining battery energy; Step 3 includes: Step 31: Given the selection strategy X m Solve the following problem to obtain the locally optimal unloading scheme L for the current iteration number. m+1 and frequency calculation scheme F m+1 : Where, μ i,j As an auxiliary variable, C i express CPU cycle count required to process a unit bit task; C j express The number of CPU cycles required to process a unit bit task; Indicates the data transfer rate during the D2D offloading phase; L j Indicates the total data size of the tasks pending execution on the SSDs; ν i Represents auxiliary variables. Indicates the data transfer rate during edge offloading; Indicates the D2D unloading phase The transmission power; Indicates edge unloading phase Transmit power when transmitting data to the BS; E i express Initial energy state; This indicates the energy consumed during the local computation phase; Step 32: Based on the locally optimal unloading scheme L obtained in step 31 m+1 and frequency calculation scheme F m+1 Calculate the locally optimal choice strategy X m+1 The solution is as follows: The above equation is solved using a solver to obtain a feasible solution; Step 33: Iterate through steps 31 and 32 until an unloading scheme, calculation frequency, and selection strategy that meet the preset conditions are obtained.

2. The method according to claim 1, characterized in that, The communication process uses orthogonal frequency division multiple access (OFDMA) technology.

3. The method according to claim 1, characterized in that, Step 3 uses the CVX solver to solve the convex optimization problem.

4. The method according to claim 1, characterized in that, The data transmission rate of the D2D unloading phase The calculation method is as follows: Where B represents the transmission bandwidth of each sub-channel. and They represent The transmit power and D2D link channel power gain, where N represents the noise power of each sub-channel.

5. The method according to claim 1, characterized in that, Data transfer rate during edge offloading The calculation formula is: Where B represents the transmission bandwidth of each sub-channel. and They represent Transmit power and cellular link channel gain when transmitting data to SSDs via D2D link, where N represents the noise power of each sub-channel.

6. A task unloading and resource allocation system, characterized in that, The system is used to implement the method as described in any one of claims 1-5, the system comprising: The system model building module is configured to build a D2D communication-assisted MEC system model, define the task processing mode of intelligent devices and the computing task offloading mode of edge servers, and divide the tasks generated by task-oriented intelligent devices into three task processing modes: local computing, D2D offloading and edge offloading. The joint optimization formula construction module is configured to jointly formulate the task partitioning strategy, computing resource allocation strategy, terminal direct transmission offloading selection strategy and dynamic pricing strategy with the optimization objective of minimizing task execution cost. The solution module is configured to decompose the joint optimization problem into multiple subproblems using the block coordinate descent method and the reconstruction-linearization technique, approximate the non-convex problem as a convex optimization problem, and apply convex optimization theory to further optimize the task offloading ratio and computational resource allocation strategy.

7. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to implement the method as described in any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 5.

Images