Task unloading method and device based on mobile cloud edge computing structure

By optimizing task offload decisions and updating service cache models in the mobile cloud edge computing structure, the problems of limited resources and large latency in mobile edge computing are solved, and efficient and low-latency task processing and energy management are achieved.

CN119030977APending Publication Date: 2024-11-26SOUTH CHINA NORMAL UNIV
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411144289.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-20
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The existing mobile edge computing model has large latency due to limited resources and long-distance transmission, which cannot meet the needs of low-latency and high-performance computing, especially in terms of service caching and computing offloading strategies. .

Method used

By obtaining the status space information, data transmission information and preset service cache model of the current time slot in the mobile cloud edge computing structure, the greedy algorithm and multi-user game Nash equilibrium method are used to optimize task unloading decisions, reduce energy consumption, and Update the service cache model through dual deep learning neural networks to adapt to dynamic environments.

Benefits of technology

It improves the resource utilization efficiency and response speed of mobile edge computing structures, is suitable for application environments with limited resources and high real-time requirements, enhances the stability and efficiency of task offloading, and reduces energy loss.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119030977A_ABST
    Figure CN119030977A_ABST
Patent Text Reader

Abstract

The invention relates to the field of task unloading, in particular to a task unloading method based on a mobile cloud edge computing structure, which comprises the following steps of: obtaining state space information of a current time slot of the mobile cloud edge computing structure, data transmission information and a preset service cache model, and inputting the state space information of the current time slot into the service cache model, action space information of the current time slot is obtained, and the action space information comprises action information corresponding to tasks requested by a plurality of Internet of Things devices; and obtaining task unloading decision information of the current time slot according to the state space information, the data transmission information and the action space information of the current time slot, and performing task unloading and execution according to the task unloading decision information of the current time slot. According to the method, the energy consumption problem faced by service caching and calculation unloading strategies in mobile edge calculation implementation is fully considered, the performance of a mobile edge calculation structure is improved through efficient resource management, energy loss is reduced, and task unloading efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of task offloading, and in particular to a task offloading method, device, computer equipment and storage medium based on a mobile cloud edge computing structure. Background Art

[0002] In the research of computation offloading of edge computing, due to the limited computing and caching resources of edge servers (ES), only some services can be cached, which directly affects the performance of computation offloading. Therefore, it is crucial to adopt an appropriate service caching strategy to reduce energy consumption and improve processing efficiency, so as to meet the latency requirements of mobile users (MUs) and optimize user experience.

[0003] With the popularity of IoT devices, the demand for low-latency and high-performance computing has increased significantly. The existing cloud computing model has large latency due to long-distance transmission and cannot meet real-time requirements. Mobile edge computing (MEC) attempts to solve this problem by providing computing resources near the user. However, in existing mobile edge computing implementations, service caching and computing offloading strategies still face problems with resource efficiency and adaptability to dynamic environments. Summary of the invention

[0004] Based on this, the purpose of the present invention is to provide a task offloading method, device, computer equipment and storage medium based on a mobile cloud edge computing structure, which fully considers the energy consumption problems faced by service caching and computing offloading strategies in the implementation of mobile edge computing, and improves the performance of the mobile edge computing structure through efficient resource management, reduces energy loss, and improves the efficiency of task offloading.

[0005] In a first aspect, an embodiment of the present application provides a task offloading method based on a mobile cloud edge computing structure, wherein the mobile cloud edge computing structure includes a plurality of IoT devices, an edge server, and a cloud service center, and the method includes the following steps:

[0006] Obtaining state space information, data transmission information, and a preset service cache model of a current time slot of a mobile cloud edge computing structure, wherein the state space information includes request task information of a plurality of IoT devices;

[0007] Inputting the state space information of the current time slot into the service cache model to obtain the action space information of the current time slot, wherein the action space information includes the action information corresponding to the tasks requested by several IoT devices;

[0008] According to the state space information, data transmission information and action space information of the current time slot, task offloading decision information of the current time slot is obtained, and according to the task offloading decision information of the current time slot, task offloading and execution are performed.

[0009] In a second aspect, an embodiment of the present application provides a task offloading device based on a mobile cloud edge computing structure, wherein the mobile cloud edge computing structure includes a plurality of IoT devices, an edge server, and a cloud service center, including:

[0010] A data acquisition module is used to obtain state space information, data transmission information and a preset service cache model of the current time slot of the mobile cloud edge computing structure, wherein the state space information includes request task information of several IoT devices;

[0011] A service cache module, used to input the state space information of the current time slot into the service cache model to obtain the action space information of the current time slot, wherein the action space information includes the action information corresponding to the tasks requested by several IoT devices;

[0012] The task unloading decision module is used to obtain the task unloading decision information of the current time slot according to the state space information, data transmission information and action space information of the current time slot, and perform task unloading and execution according to the task unloading decision information of the current time slot.

[0013] In a third aspect, an embodiment of the present application provides a computer device comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor; when the computer program is executed by the processor, the steps of the task offloading method based on the mobile cloud edge computing structure as described in the first aspect are implemented.

[0014] In a fourth aspect, an embodiment of the present application provides a storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the task offloading method based on the mobile cloud edge computing structure as described in the first aspect are implemented.

[0015] In an embodiment of the present application, a task offloading method, apparatus, computer equipment and storage medium based on a mobile cloud edge computing structure are provided, which fully consider the energy consumption problems faced by service caching and computing offloading strategies in the implementation of mobile edge computing, and improve the performance of the mobile edge computing structure through efficient resource management, reduce energy loss, and improve the efficiency of task offloading.

[0016] For better understanding and implementation, the present invention is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a task offloading method based on a mobile cloud edge computing structure provided by an embodiment of the present application;

[0018] Figure 2A schematic diagram of the process of S2 in a task offloading method based on a mobile cloud edge computing structure provided in one embodiment of the present application;

[0019] Figure 3 A schematic diagram of the process of S3 in a task offloading method based on a mobile cloud edge computing structure provided in one embodiment of the present application;

[0020] Figure 4 A schematic diagram of the process of S4 in a task offloading method based on a mobile cloud edge computing structure provided in another embodiment of the present application;

[0021] Figure 5 A schematic diagram of the process of S41 in a task offloading method based on a mobile cloud edge computing structure provided in one embodiment of the present application;

[0022] Figure 6 A schematic diagram of the process of S42 in a task offloading method based on a mobile cloud edge computing structure provided in one embodiment of the present application;

[0023] Figure 7 A schematic diagram of the process of S422 in a task offloading method based on a mobile cloud edge computing structure provided in one embodiment of the present application;

[0024] Figure 8 A schematic diagram of the structure of a task offloading device based on a mobile cloud edge computing structure provided by an embodiment of the present application;

[0025] Fig. 9 A schematic diagram of the structure of a computer device provided for one embodiment of the present application. DETAILED DESCRIPTION

[0026] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0027] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0028] It should be understood that, although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" / "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determination".

[0029] The executor of the task offloading method based on the mobile cloud edge computing structure is the offloading device of the task offloading method based on the mobile cloud edge computing structure (hereinafter referred to as the offloading device). The offloading device can be implemented by software and / or hardware, and the task offloading method based on the mobile cloud edge computing structure can be implemented by software and / or hardware. The offloading device can be composed of two or more physical entities, or it can be composed of one physical entity. The hardware pointed to by the offloading device essentially refers to computer equipment. For example, the offloading device can be a computer, a mobile phone, a tablet or an interactive tablet and other devices. In an optional embodiment, the offloading device can specifically be a server, or a server cluster composed of multiple computer devices.

[0030] Computation offloading and service caching in a three-layer mobile cloud edge computing architecture. Services are dynamically cached on edge servers (ES) to assist mobile users (MUs) in task execution. With cached services, mobile users can complete tasks through local user computing, edge server computing offloading, and cloud service center (CSC) computing offloading.

[0031] The application environment of the task offloading method based on the mobile cloud edge computing structure provided by an embodiment of the present application is a mobile cloud edge computing structure. The mobile cloud edge computing structure includes an Internet of Things network, an edge server and a cloud computing server. The Internet of Things network includes a number of Internet of Things devices, and the several Internet of Things devices are communicated with the edge server, and the edge server and the cloud computing server are communicated with each other.

[0032] The IoT device is an IoT device carried by a mobile user. The edge server provides dynamic cache services and edge computing services for the IoT device to assist the task execution of the IoT device. The cloud computing server provides cloud computing services for the IoT device. With the help of cache services, the IoT device can complete the task through local user computing, edge server computing offloading, and cloud service center computing offloading.

[0033] See also Figure 1 , Figure 1A flowchart of a task offloading method based on a mobile cloud edge computing structure provided in one embodiment of the present application is provided, wherein the method comprises the following steps:

[0034] S1: Obtain the state space information, data transmission information, and preset service cache model of the current time slot of the mobile cloud edge computing structure.

[0035] In this embodiment, the offloading device obtains the state space information of the current time slot of the mobile cloud edge computing structure, wherein the state space information includes the request task information of several IoT devices, and the request task information is used to indicate the task requested by the IoT device, and the state space information is:

[0036]

[0037] In the formula, S t is the state space information of the tth time slot, is the request task information of the Uth IoT device in the state space information of the tth time slot, represents the request task k of the Uth IoT device. Specifically, each IoT device request task is a task.

[0038] The offloading device obtains data transmission information and a preset service cache model, wherein the data transmission information includes channel bandwidth parameters, transmission power parameters and channel gain parameters of several IoT devices, the channel bandwidth parameters include bandwidth parameters of several channels, and the channel gain parameters include gain parameters of several channels. The service cache model is used to output service cache space information, wherein the service cache space information includes service cache information of tasks requested by several IoT devices, and the service cache information is used to indicate whether the edge server has cached services corresponding to the tasks.

[0039] S2: Input the state space information of the current time slot into the service cache model to obtain the action space information of the current time slot.

[0040] In this embodiment, the offloading device inputs the state space information of the current time slot into the service cache model to obtain the action space information of the current time slot, wherein the action space information includes action information corresponding to tasks requested by several IoT devices.

[0041] See also Figure 2 , Figure 2 The flowchart of S2 in the task offloading method based on the mobile cloud edge computing structure provided in one embodiment of the present application includes step S21, which is as follows:

[0042] S21: According to the state space information of the current time slot and the policy network, a greedy algorithm is used to obtain the service cache space information of the next time slot as the action space information of the current time slot.

[0043] In this embodiment, the unloading device uses a greedy algorithm to randomly select an action according to the state space information of the current time slot and the policy network, that is, the probability of randomly caching the service is δ or the probability of selecting the action with the highest Q value through Q value calculation is 1-δ, and obtains the service cache space information of the next time slot as the action space information of the current time slot, which is specifically as follows:

[0044] a t =β t+1

[0045]

[0046] In the formula, a t is the action space information of the t-th time slot, is the action information corresponding to the kth task in the action space information of the tth time slot, β t+1 Cache spatial information for the service in the t+1th time slot, The service cache information corresponding to the i-th task in the t+1-th time slot, Indicates that the service corresponding to the task will be cached. This means that the service corresponding to the task is not cached.

[0047] S3: Obtain task offloading decision information of the current time slot according to the state space information, data transmission information and action space information of the current time slot, and perform task offloading and execution according to the task offloading decision information of the current time slot.

[0048] In this embodiment, the unloading device obtains the task unloading decision information of the current time slot according to the state space information, data transmission information and action space information of the current time slot, and performs task unloading and execution according to the task unloading decision information of the current time slot.

[0049] See also Figure 3 , Figure 3 The flowchart of S3 in the task offloading method based on the mobile cloud edge computing structure provided in one embodiment of the present application includes steps S31 to S32, which are as follows:

[0050] S31: According to the data transmission information and a preset transmission rate calculation algorithm, the transmission rate between the several IoT devices and the edge server is obtained, and according to the transmission rate of the several IoT devices, the state space information and the action space information, the strategy space of the tasks requested by the several IoT devices is obtained.

[0051] In this embodiment, the unloading device obtains the transmission rate between the plurality of IoT devices and the edge server according to the data transmission information and a preset transmission rate calculation algorithm, wherein the transmission rate calculation algorithm is:

[0052]

[0053] In the formula, r u,e is the transmission rate between the u-th IoT device and the edge server, e represents the edge server, is the bandwidth parameter of the nth channel of the uth IoT device, q u is the transmission power parameter of the u-th IoT device, is the gain parameter of the nth channel of the uth IoT device, σ is the noise background power parameter, v∈U\{u} is the other IoT devices in the IoT network, U is the total number of IoT devices in the IoT network, U\{u} represents the set of other IoT devices in the IoT network except the uth IoT device, q v is the transmission power parameter of the vth IoT device, is the gain parameter between the vth IoT device and the edge server.

[0054] The offloading device independently selects the optimal computing decision for each IoT device according to the transmission rates of the IoT devices, the state space information, and the action space information, so as to minimize its energy consumption and obtain the strategy space of the tasks requested by the IoT devices.

[0055] S32: Construct an energy consumption function of the IoT device, and based on the energy consumption function of the IoT device, adopt a multi-user game Nash equilibrium method, with the goal of minimizing the energy consumption function of the IoT device, extract corresponding task offloading indication information from the strategy space of tasks requested by several IoT devices, and obtain the task offloading decision information.

[0056] In this embodiment, the uninstall device constructs an IoT device energy consumption function, wherein the IoT device energy consumption function is:

[0057]

[0058] In the formula, is the energy consumption function of IoT devices, η t is the task offloading decision information for the tth time slot, η u,t is the task offloading instruction information of the u-th IoT device in the t-th time slot, η v,t It is the task offloading instruction information of the vth IoT device in the tth time slot.

[0059] The offloading device adopts a multi-user game Nash equilibrium method according to the energy consumption function of the Internet of Things device, with the goal of minimizing the energy consumption function of the Internet of Things device. While ensuring that the local computing cost of the task requested by the Internet of Things device is greater than the task offloading cost, the Internet of Things device u will give priority to offloading the task to the edge server for task calculation, and when the edge server does not have a service corresponding to the cached task, the edge server will offload the task to the cloud service center for processing, while ensuring that each user can efficiently make task offloading decisions, minimize the energy consumption of the Internet of Things device, obtain the optimal task offloading strategy of the mobile cloud edge computing structure and achieve Nash equilibrium, extract corresponding task offloading indication information from the policy space of tasks requested by several Internet of Things devices, and obtain the task offloading decision information, wherein the task offloading indication information is used to instruct the Internet of Things device to offload the corresponding task to the edge server or cloud service center.

[0060] By modeling the task offloading problem as a multi-user game and using a decentralized algorithm to solve its Nash equilibrium solution, it not only improves the resource utilization efficiency and response speed of the mobile cloud edge computing structure, but also is suitable for application environments with limited resources and high real-time requirements. It also enhances the adaptability of the mobile cloud edge computing structure to demand changes and improves the stability and efficiency of task offloading.

[0061] In an optional embodiment, the step S4 is also included: updating the model parameters of the service cache model. Figure 4 , Figure 4 A flow chart of S4 in a task offloading method based on a mobile cloud edge computing structure provided in another embodiment of the present application includes steps S41 to S42, which are specifically as follows:

[0062] S41: After task unloading and execution, obtain the energy consumption information of the current time slot, use the energy consumption information as the reward information, and obtain the reward information of the current time slot; obtain the state space information of the next time slot, combine the state space information, action space information, reward information of the current time slot and the state space information of the next time slot, and construct a training information combination for the current time slot.

[0063] In this embodiment, the unloading device unloads and executes the task according to the task unloading decision information of the current time slot, obtains the energy consumption information of the current time slot, uses the energy consumption information as the reward information, and obtains the reward information of the current time slot.

[0064] The offloading device obtains the state space information of the next time slot, combines the state space information, action space information, reward information of the current time slot and the state space information of the next time slot, and constructs the training information combination of the current time slot to update the service cache model. It fully considers the energy consumption problems faced by service cache and computing offloading strategies in the implementation of mobile edge computing, reduces energy loss, and improves the dynamic environment adaptability and processing efficiency of the mobile edge computing structure.

[0065] In an optional embodiment, in order to prevent the algorithm from falling into a local optimal solution during training and to better explore a non-stationary environment, the offloading device adds Gaussian noise to the action space information.

[0066] See also Figure 5 , Figure 5 The flowchart of S41 in the task offloading method based on the mobile cloud edge computing structure provided in one embodiment of the present application includes steps S411 to S413, which are as follows:

[0067] S411: If the task is executed locally, obtain the computing task load parameters and the computing task capability parameters of the IoT device, and obtain the first energy consumption information of the current time slot according to a preset first energy consumption calculation algorithm.

[0068] If the task is executed locally, in this embodiment, the offloading device obtains the computing task load parameter and the computing task capability parameter of the IoT device, and obtains the first energy consumption information of the current time slot according to the preset first energy consumption calculation algorithm, wherein the first energy consumption information includes energy consumption data of tasks requested by several IoT devices executed locally, and the first energy consumption calculation algorithm is:

[0069]

[0070] In the formula, E u,l is the energy consumption data of the task requested by the u-th IoT device when executed locally, l is the local execution identifier, is the energy consumption function, f l,k is the computing task capability parameter of the u-th IoT device to perform the k-th task, s k is the calculation task load parameter of the kth task.

[0071] S412: If the task is offloaded to the edge server and the edge server does not cache the corresponding service of the task, obtain the task input parameters, task transmission power and the transmission rate between the edge server and the cloud service center, and obtain the second energy consumption information of the current time slot according to the preset second energy consumption calculation algorithm.

[0072] If the task is offloaded to the edge server, and the edge server does not cache the corresponding service of the task, it means that the edge server needs to offload the task to the cloud service center again for execution in the cloud service center. In this embodiment, the offloading device obtains the task input parameters, the task transmission power, and the transmission rate between the edge server and the cloud service center, and obtains the second energy consumption information of the current time slot according to the preset second energy consumption calculation algorithm, wherein the second energy consumption information includes the energy consumption data of the tasks requested by several IoT devices executed in the cloud service center, and the second energy consumption calculation algorithm is:

[0073]

[0074] In the formula, E u,c is the energy consumption data of the task requested by the u-th IoT device executed in the cloud service center, c is the execution identifier of the cloud service center, p k is the task transmission rate of the kth task, I k The task input parameters for the kth task, r e,c is the transmission rate between the edge server and the cloud service center.

[0075] S413: If the task is offloaded to the edge server, and the edge server has a service corresponding to the task cached, the task input parameter and the task transmission power are obtained, and the third energy consumption information of the current time slot is obtained according to a preset third energy consumption calculation algorithm.

[0076] If the task is offloaded to the edge server, and the edge server has a service corresponding to the task cached, it means that the task requested by the IoT device is executed in the edge server. In this embodiment, the offloading device obtains the task input parameters and the task transmission power, and obtains the third energy consumption information of the current time slot according to the preset third energy consumption calculation algorithm, wherein the third energy consumption information includes energy consumption data of tasks requested by several IoT devices executed in the edge server, and the third energy consumption calculation algorithm is:

[0077]

[0078] In the formula, E u,e Energy consumption data for the task requested by the u-th IoT device executed on the edge server.

[0079] S42: Repeatedly construct training information combinations to obtain training information combinations of several time slots, and update the model parameters of the service cache model according to the training information combinations of several time slots, so as to obtain the action space information of the next time slot through the state space information of the next time slot through the updated service cache model.

[0080] In this embodiment, the offloading device repeatedly constructs the training information combination to obtain the training information combination of several time slots, updates the model parameters of the service cache model according to the training information combination of several time slots, and obtains the target service cache model.

[0081] In order to improve the stability and convergence of the model training process, thereby improving the learning efficiency and the stability of the model, specifically, the unloading device adopts an experience replay mechanism, repeats steps S1 to S3 and step S4, constructs initial training information combinations for several time slots, and stores them in a preset database. The unloading device obtains initial training information combinations for several time slots by randomly sampling the initial training information combinations of the minimum batch. The unloading device updates the model parameters of the service cache model according to the training information combinations of several time slots, so as to obtain the action space information of the next time slot through the updated service cache model for the state space information of the next time slot, so as to execute in the environment and obtain the maximum long-term return.

[0082] The service cache model adopts a dual deep learning neural network (Double-Dueling DQN), including a training network and a target network. Figure 6 , Figure 6 The flowchart of S42 in the task offloading method based on the mobile cloud edge computing structure provided in one embodiment of the present application includes steps S421 to S425, which are as follows:

[0083] S421: Input the state space information and action space information of the current time slot in the training information combination into the training network, and obtain the action value information of the current time slot according to a preset action value calculation algorithm.

[0084] In this embodiment, the unloading device inputs the state space information and action space information of the current time slot in the training information combination into the training network, and obtains the action value information of the current time slot according to the preset action value calculation algorithm, wherein the action value information includes the action value parameters of the task actions requested by several IoT devices; the service value calculation algorithm is:

[0085]

[0086] In the formula, q(S t ,a t ) is the action value information of the tth time slot, S t is the state space information of the tth time slot, a t is the action space information of the t-th time slot, is the service action information corresponding to the kth task in the action space information of the tth time slot; V(·) is the state value function, is the action advantage function, The total number of actions to choose, a′ t Select an action from the total number of action choices for the tth time slot.

[0087] S422: Obtain service memory space information of the current time slot, and obtain service cache space information of the next time slot using a dynamic programming method according to the service memory space information of the current time slot and the action value information of the current time slot.

[0088] In this embodiment, the offloading device obtains the service memory space information of the current time slot, wherein the service memory space information includes service memory capacity information corresponding to tasks requested by several IoT devices.

[0089] The unloading device obtains the service cache space information of the next time slot by a dynamic programming method according to the service memory space information of the current time slot and the action value information of the current time slot, wherein the service cache space information includes the service cache space information corresponding to the tasks requested by several IoT devices.

[0090] See also Figure 7 , Figure 7 The flowchart of S422 in the task offloading method based on the mobile cloud edge computing structure provided in one embodiment of the present application includes steps S4221 to S4222, which are as follows:

[0091] S4221: Use dynamic programming method to construct service cache decision matrix and maximum cache value matrix.

[0092] In this embodiment, the offloading device adopts a dynamic programming method to construct a service cache decision matrix and a maximum cache value matrix, wherein the service cache decision matrix is ​​used to record the service cache decisions of tasks requested by several IoT devices under different current memory capacities of the edge server; the maximum cache value matrix is ​​used to record the maximum cache values ​​of tasks requested by several IoT devices under different current memory capacities of the edge server, and the service cache decision matrix is:

[0093]

[0094] In the formula, Φ r (k,z) is the service cache decision matrix, z is the current memory capacity of the edge server, k is the kth task, which means the maximum cache value of the task requested by the kth IoT device when the current memory capacity of the edge server is z, Φ v (k-1,z-fD k ) indicates that the current memory capacity of the edge server is z-fDk Next, the service cache decision for the task requested by the k-1th IoT device, argmax f∈{0,1} () is the aggregate function, f is the cache identifier, D k is the memory capacity of the kth task.

[0095] The maximum cache value matrix is:

[0096]

[0097] In the formula, Φ v (k,z) is the maximum cache value matrix, which represents the maximum cache value of the task requested by the kth IoT device when the current memory capacity of the edge server is z.

[0098] Specifically, the offloading device introduces a K×Z matrix Φ to construct an initial service cache decision matrix and an initial maximum cache value matrix, respectively, where K is the total number of tasks, Z is the total memory capacity of the edge server, and the initial maximum cache value matrix is ​​Φ v (K,Z)=[0] K×z , the initial service cache decision matrix is ​​Φ r (K,Z)=[0] K×Z .

[0099] The unloading device traverses the services from 1 to K-1 through the outer loop and traverses the edge server's current memory capacity z from 0 to Z through the inner loop according to the service memory space information of the current time slot and the action value information of the current time slot. For the service corresponding to the first task, if the service corresponding to the current task is the first service and the edge server's current memory capacity z is less than the memory occupied by the current service 1, that is, the memory capacity of service 1, then Φ r (k,z) and Φ v (k,z) is set to 0, indicating that the service corresponding to the task is not cached; otherwise, Φ r (k,z) is set to 1, and q(S t ,a k ) is assigned to Φ v (k,z).

[0100] For the subsequent services corresponding to the 2nd to K-1th tasks, if the current memory capacity z of the edge server is less than D k , D k is the memory capacity of the kth task, that is, the current memory capacity z of the edge server is less than the memory occupied by the service of the current kth task, then Φ r (k,z) is set to 0, Φ v (k,z)=Φ v (k-1,z).

[0101] If the edge server's current memory capacity z is greater than D k , then we need to compare Φ v (k-1,z) and Φ v (k-1,zD k )+Φ r (k,z)q(S t ,a k ) size, if Φ v (k-1,z)<Φ v (k-1,zD k )+Φ r (k,z)q(S t ,a k ), it means that the value of the best combination of services corresponding to the first k tasks is greater than the value of the best combination of services corresponding to k-1 tasks, and the service is cached, and Φ v (k,z)=Φ v (k-1,z-fD k )+Φ r (k,z)q(S t ,a k ), Φ r (k,z)=1; otherwise, the service is not cached and Φ v (k,z)=Φ v (k-1,z),Φ r (k,z)=0; repeat the above operation until k=K-1.

[0102] For the service corresponding to the last Kth task, only the comparison is performed when z=Z, and there is no need to perform an outer loop to traverse the service and an inner loop to traverse the edge server. k , then Φ r (K, Z) is set to 0, Φ v (K,z)=Φ v (K-1,z).

[0103] If the edge server's current memory capacity z is greater than D K , then we need to compare Φ v (K-1,z) and Φ v (K-1,zD K )+Φ r (K,z)q(S t ,a K ) size, if Φ v (K-1,z)<Φ v (K-1,zD K )+Φ r (K,z)q(S t ,aK ), it means that the value of the best combination of services corresponding to the first K tasks is greater than the value of the best combination of services corresponding to K-1 tasks, and the service is cached, and Φ v (K,z)=Φ v (K-1,z-fD K )+Φ r (K,z)q(S t ,a K ), Φ r (K,z)=1; otherwise, the service is not cached and Φ v (K,z)=Φ v (K-1,z),Φ r (K,z)=0.

[0104] S4222: construct an algorithm based on the service cache decision matrix, the maximum cache value matrix and the preset service cache space information to obtain the service cache space information of the next time slot.

[0105] The service cache space information construction algorithm is:

[0106]

[0107] In the formula, is the service cache information of the kth task in the service cache space information of the t+1th time slot, and K is the total number of tasks.

[0108] In this embodiment, the offloading device constructs an algorithm based on the service cache decision matrix, the maximum cache value matrix and the preset service cache space information, and traverses backward from K to 1 to obtain the service cache space information of the next time slot.

[0109] S423: Obtain the evaluation value parameter of the current time slot according to the action value information of the current time slot, the optimal service cache space information of the next time slot and the preset evaluation value calculation algorithm.

[0110] The evaluation value calculation algorithm is:

[0111]

[0112] In the formula, y t is the evaluation value parameter of the tth time slot, U is the total number of tasks, The optimal service cache corresponding to the i-th task in the t+1-th time slot caches the information, It is the service action information corresponding to the i-th task in the action space information of the t-th time slot.

[0113] In this embodiment, the offloading device obtains the evaluation value parameter of the current time slot according to the action value information of the current time slot, the optimal service cache space information of the next time slot, and a preset evaluation value calculation algorithm.

[0114] S424: According to the state space information of the next time slot and the strategy network, a greedy algorithm is used to obtain the service cache space information of the next two time slots as the action space information of the next time slot, and the state space information, action space information of the next time slot and the reward information of the current time slot are input into the target network, and the target value parameters of the current time slot are obtained according to the preset target value calculation algorithm.

[0115] In this embodiment, the offloading device uses a greedy algorithm to obtain the service cache space information of the next two time slots according to the state space information of the next time slot and the policy network as the action space information of the next time slot. For the specific embodiment, please refer to step S21, which will not be repeated here.

[0116] The unloading device inputs the state space information, action space information of the next time slot and the reward information of the current time slot into the target network, and obtains the target value parameter of the current time slot according to the preset target value calculation algorithm, wherein the target value calculation algorithm is:

[0117] y′ t =r t +γQ′(S t+1 ,a t+1 )|θ Q′ )

[0118] In the formula, y′ t is the target value parameter of the tth time slot, r t is the reward information of the tth time slot, S t+1 is the state space information of the t+1th time slot, a t+1 is the action space information of the t+1th time slot, Q′() is the target value network function, θ Q′ are the network parameters of the target value network.

[0119] S425: Perform loss calculation according to the evaluation value parameter and the target value parameter of the current time slot, update the parameters of the training network according to the obtained loss value, obtain an updated training network, and adopt a soft update strategy to update the parameters of the updated training network to the target network to obtain an updated target network.

[0120] In this embodiment, the unloading device performs loss calculation according to the evaluation value parameter and the target value parameter of the current time slot, updates the parameters of the training network according to the obtained loss value, obtains an updated training network, and adopts a soft update strategy to update the parameters of the updated training network to the target network to obtain an updated target network. Through the above steps, the training network continuously improves its estimate of the Q value, thereby providing more accurate feedback to the policy network and helping the policy network optimize its strategy. The energy consumption problem faced by the service cache and computing offloading strategy in the implementation of mobile edge computing is fully considered to update the service cache model so that the service cache model can adapt to the dynamically changing system environment, improve the performance of the mobile edge computing structure through efficient resource management, reduce energy loss, and improve the efficiency of task offloading.

[0121] Please refer to Figure 8 , Figure 8 A schematic diagram of a task offloading device based on a mobile cloud edge computing structure provided in one embodiment of the present application. The device can implement all or part of the task offloading device based on a mobile cloud edge computing structure through software, hardware, or a combination of both. The device 8 includes:

[0122] A data acquisition module 81 is used to obtain state space information, data transmission information, and a preset service cache model of a current time slot of a mobile cloud edge computing structure, wherein the state space information includes request task information of a plurality of IoT devices;

[0123] The service cache module 82 is used to input the state space information of the current time slot into the service cache model to obtain the action space information of the current time slot, wherein the action space information includes the action information corresponding to the tasks requested by the multiple IoT devices;

[0124] The task offloading decision module 83 is used to obtain the task offloading decision information of the current time slot according to the state space information, data transmission information and action space information of the current time slot, and perform task offloading and execution according to the task offloading decision information of the current time slot.

[0125] In the embodiment of the present application, the state space information, data transmission information and preset service cache model of the current time slot of the mobile cloud edge computing structure are obtained through the data acquisition module, wherein the state space information includes the request task information of several IoT devices; the state space information of the current time slot is input into the service cache model through the service cache module to obtain the action space information of the current time slot, wherein the action space information includes the action information corresponding to the tasks requested by several IoT devices; the task unloading decision information of the current time slot is obtained according to the state space information, data transmission information and action space information of the current time slot through the task unloading decision module, and the task is unloaded and executed according to the task unloading decision information of the current time slot. The energy consumption problem faced by the service cache and computing unloading strategy in the implementation of mobile edge computing is fully considered, and the service cache model is used to update the service cache model so that the service cache model can adapt to the dynamically changing system environment, improve the performance of the mobile edge computing structure through efficient resource management, reduce energy loss, and improve the efficiency of task unloading.

[0126] Please refer to Fig. 9 , Fig. 9 The computer device 9 is a schematic diagram of a structure of a computer device provided in an embodiment of the present application. The computer device 9 includes: a processor 91, a memory 92, and a computer program 93 stored in the memory 92 and executable on the processor 91; the computer device may store a plurality of instructions, which are suitable for being loaded and executed by the processor 91 Figures 1 to 7 The method steps shown in the figure can be found in the specific execution process. Figures 1 to 7 The specific description shown will not be repeated here.

[0127] Among them, the processor 91 may include one or more processing cores. The processor 91 uses various interfaces and lines to connect various parts in the server, and executes various functions and processes data of the task offloading device 8 based on the mobile cloud edge computing structure by running or executing instructions, programs, code sets or instruction sets stored in the memory 92, and calling the data in the memory 92. Optionally, the processor 91 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programble Logic Array, PLA). The processor 91 can integrate one or more combinations of a central processing unit 91 (Central Processing Unit, CPU), an image processor 91 (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the touch display; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 91, and it can be implemented by a single chip.

[0128] Among them, the memory 92 may include a random access memory 92 (Random Access Memory, RAM), and may also include a read-only memory 92 (Read-Only Memory). Optionally, the memory 92 includes a non-transitory computer-readable storage medium. The memory 92 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 92 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch instructions, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 92 may also be optionally at least one storage device located away from the aforementioned processor 91.

[0129] The present application also provides a storage medium that can store multiple instructions, which are suitable for the processor to load and execute the above-mentioned Figures 1 to 7 The method steps shown in the figure can be found in the specific execution process. Figures 1 to 7 The specific description shown will not be repeated here.

[0130] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0131] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0132] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraint algorithm of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0133] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0134] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0135] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0136] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc.

[0137] The present invention is not limited to the above-mentioned embodiments. If various changes or modifications to the present invention do not depart from the spirit and scope of the present invention, and if these changes and modifications fall within the scope of the claims and equivalent technologies of the present invention, the present invention is also intended to include these changes and modifications.

Claims

1. A task offloading method based on a mobile cloud edge computing structure, wherein the mobile cloud edge computing structure includes a plurality of Internet of Things devices, an edge server and a cloud service center, characterized in that: The method comprises the following steps: Obtaining state space information, data transmission information, and a preset service cache model of a current time slot of a mobile cloud edge computing structure, wherein the state space information includes request task information of a plurality of IoT devices; Inputting the state space information of the current time slot into the service cache model to obtain the action space information of the current time slot, wherein the action space information includes the action information corresponding to the tasks requested by several IoT devices; According to the state space information, data transmission information and action space information of the current time slot, task offloading decision information of the current time slot is obtained, and according to the task offloading decision information of the current time slot, task offloading and execution are performed.

2. The task offloading method based on the mobile cloud edge computing structure according to claim 1 is characterized in that: The requested task information is used to indicate the task requested by the IoT device; the service cache model includes a policy network; The step of inputting the state space information of the current time slot into the service cache model to obtain the action space information of the current time slot of the Internet of Things network comprises the following steps: According to the state space information of the current time slot and the policy network, a greedy algorithm is used to obtain the service cache space information of the next time slot as the action space information of the current time slot, wherein the service cache space information includes the service cache information of tasks requested by several IoT devices, and the service cache information is used to indicate whether the edge server has cached services corresponding to the tasks.

3. The task offloading method based on the mobile cloud edge computing structure according to claim 2 is characterized in that: The data transmission information includes channel bandwidth parameters, transmission power parameters and channel gain parameters of several IoT devices, wherein the channel bandwidth parameters include bandwidth parameters of several channels, and the channel gain parameters include gain parameters of several channels; The step of obtaining task offloading decision information of the current time slot according to the state space information, data transmission information and action space information of the current time slot comprises the following steps: According to the data transmission information and the preset transmission rate calculation algorithm, the transmission rate between the several IoT devices and the edge server is obtained, and according to the transmission rate of the several IoT devices, the state space information and the action space information, the strategy space of the tasks requested by the several IoT devices is obtained, wherein the transmission rate calculation algorithm is: In the formula, r u,e is the transmission rate between the u-th IoT device and the edge server, e represents the edge server, is the bandwidth parameter of the nth channel of the uth IoT device, q u is the transmission power parameter of the u-th IoT device, is the gain parameter of the nth channel of the uth IoT device, σ is the noise background power parameter, v∈U\{u} is the other IoT devices in the IoT network, U is the total number of IoT devices in the IoT network, U\{u} represents the set of other IoT devices in the IoT network except the uth IoT device, q v is the transmission power parameter of the vth IoT device, is the gain parameter between the vth IoT device and the edge server; An energy consumption function of an IoT device is constructed. According to the energy consumption function of the IoT device, a multi-user game Nash equilibrium method is adopted to minimize the energy consumption function of the IoT device. Corresponding task offloading indication information is extracted from the strategy space of tasks requested by several IoT devices to obtain the task offloading decision information, wherein the task offloading indication information is used to instruct the IoT device to offload the corresponding task to the edge server or cloud service center. The energy consumption function of the IoT device is: In the formula, is the energy consumption function of IoT devices, η t is the task offloading decision information for the tth time slot, η u,t is the task offloading instruction information of the u-th IoT device in the t-th time slot, η v,t It is the task offloading instruction information of the vth IoT device in the tth time slot.

4. The task offloading method based on mobile cloud edge computing structure according to claim 3 is characterized in that: The step of performing task unloading and execution according to the task unloading decision information of the current time slot to obtain the energy consumption information of the current time slot includes the following steps: If the task is executed locally, the computing task load parameter and the computing task capability parameter of the IoT device are obtained, and the first energy consumption information of the current time slot is obtained according to a preset first energy consumption calculation algorithm, wherein the first energy consumption information includes energy consumption data of tasks requested by several IoT devices executed locally, and the first energy consumption calculation algorithm is: In the formula, E u,l is the energy consumption data of the task requested by the u-th IoT device when executed locally, l is the local execution identifier, is the energy consumption function, f l,k is the computing task capability parameter of the u-th IoT device to perform the k-th task, s k is the calculation task load parameter of the kth task; If the task is offloaded to the edge server, and the edge server does not cache the corresponding service of the task, the task input parameters, the task transmission power, and the transmission rate between the edge server and the cloud service center are obtained, and the second energy consumption information of the current time slot is obtained according to the preset second energy consumption calculation algorithm, wherein the second energy consumption information includes the energy consumption data of the tasks requested by several IoT devices executed in the cloud service center, and the second energy consumption calculation algorithm is: In the formula, E u,c is the energy consumption data of the task requested by the u-th IoT device executed in the cloud service center, c is the execution identifier of the cloud service center, p k is the task transmission rate of the kth task, I k The task input parameters for the kth task, r e,c is the transmission rate between the edge server and the cloud service center; If the task is unloaded to the edge server, and the edge server has a service corresponding to the task in the cache, the task input parameters and the task transmission power are obtained, and the third energy consumption information of the current time slot is obtained according to the preset third energy consumption calculation algorithm, wherein the third energy consumption information includes energy consumption data of tasks requested by several IoT devices executed on the edge server, and the third energy consumption calculation algorithm is: In the formula, E u,e Energy consumption data of the task requested by the u-th IoT device executed on the edge server.

5. The task offloading method based on the mobile cloud edge computing structure according to any one of claims 2 to 4, characterized in that: It also includes the steps of: updating the model parameters of the service cache model; The updating of the model parameters of the service cache model comprises the steps of: After unloading and executing the task, energy consumption information of the current time slot is obtained, and the energy consumption information is used as the reward information to obtain reward information of the current time slot; Obtain the state space information of the next time slot, combine the state space information, action space information, reward information of the current time slot and the state space information of the next time slot, and construct the training information combination of the current time slot; Repeatedly construct training information combinations to obtain training information combinations for several time slots, and update the model parameters of the service cache model based on the training information combinations for several time slots, so as to obtain the action space information for the next time slot through the updated service cache model for the state space information of the next time slot.

6. The task offloading method based on the mobile cloud edge computing structure according to claim 5 is characterized in that: The service cache model adopts a dual deep learning neural network, including a training network and a target network; The method of updating the model parameters of the service cache model according to the combination of training information of several time slots to obtain the target service cache model for performing the service cache operation comprises the steps of: The state space information and action space information of the current time slot in the training information combination are input into the training network, and the action value information of the current time slot is obtained according to the preset action value calculation algorithm, wherein the action value information includes the action value parameters of the task actions requested by several IoT devices; the service value calculation algorithm is: In the formula, q(S t ,a t ) is the action value information of the tth time slot, S t is the state space information of the tth time slot, a t is the action space information of the t-th time slot, is the action information corresponding to the kth task in the action space information of the tth time slot; V(·) is the state value function, is the action advantage function, The total number of actions to choose, a′ t Select an action from the total number of action choices for the tth time slot; Obtain service memory space information of the current time slot, and obtain service cache space information of the next time slot using a dynamic programming method according to the service memory space information of the current time slot and the action value information of the current time slot, wherein the service memory space information includes service memory capacity information corresponding to tasks requested by several IoT devices, and the service cache space information includes service cache cache information corresponding to tasks requested by several IoT devices; According to the action value information of the current time slot, the optimal service cache space information of the next time slot and the preset evaluation value calculation algorithm, the evaluation value parameter of the current time slot is obtained, wherein the evaluation value calculation algorithm is: In the formula, y t is the evaluation value parameter of the tth time slot, U is the total number of tasks, The service cache information corresponding to the i-th task in the t+1-th time slot, is the action information corresponding to the i-th task in the action space information of the t-th time slot; According to the state space information of the next time slot and the policy network, a greedy algorithm is used to obtain the service cache space information of the next two time slots as the action space information of the next time slot, and the state space information, action space information of the next time slot and the reward information of the current time slot are input into the target network, and the target value parameter of the current time slot is obtained according to the preset target value calculation algorithm, wherein the target value calculation algorithm is: y′ t =r t +γQ′(S t+1 ,a t+1 )|θ Q′ ) In the formula, y′ t is the target value parameter of the tth time slot, r t is the reward information of the tth time slot, S t+1 is the state space information of the t+1th time slot, a t+1 is the action space information of the t+1th time slot, Q′() is the target value network function, θ Q′ is the network parameter of the target value network; According to the evaluation value parameter and the target value parameter of the current time slot, the loss calculation is performed, and according to the obtained loss value, the parameters of the training network are updated to obtain an updated training network, and a soft update strategy is adopted to update the parameters of the updated training network to the target network to obtain an updated target network.

7. The task offloading method based on mobile cloud edge computing structure according to claim 6 is characterized in that: The method of obtaining the service cache space information of the next time slot by using a dynamic programming method according to the service memory space information of the current time slot and the action value information of the current time slot comprises the following steps: A dynamic programming method is used to construct a service cache decision matrix and a maximum cache value matrix, wherein the service cache decision matrix is ​​used to record the service cache decisions of tasks requested by several IoT devices under different current memory capacities of the edge server; the maximum cache value matrix is ​​used to record the maximum cache values ​​of tasks requested by several IoT devices under different current memory capacities of the edge server. The service cache decision matrix is: In the formula, Φ r (k,z) is the service cache decision matrix, z is the current memory capacity of the edge server, k is the kth task, which means the maximum cache value of the task requested by the kth IoT device when the current memory capacity of the edge server is z, Φ v (k-1,z-fD k ) indicates that the current memory capacity of the edge server is z-fD k Next, the service cache decision for the task requested by the k-1th IoT device, argmax f∈{0,1} () is the aggregate function, f is the cache identifier, D k is the memory capacity of the kth task; The maximum cache value matrix is: In the formula, Φ v (k,z) is the maximum cache value matrix, which indicates the maximum cache value of the task requested by the kth IoT device when the current memory capacity of the edge server is z; According to the service cache decision matrix, the maximum cache value matrix and the preset service cache space information construction algorithm, the service cache space information of the next time slot is obtained, wherein the service cache space information construction algorithm is: In the formula, is the service cache information of the kth task in the service cache space information of the t+1th time slot, and K is the total number of tasks.

8. A task offloading device based on a mobile cloud edge computing structure, wherein the mobile cloud edge computing structure includes a plurality of Internet of Things devices, an edge server and a cloud service center, characterized in that: include: A data acquisition module is used to obtain state space information, data transmission information and a preset service cache model of the current time slot of the mobile cloud edge computing structure, wherein the state space information includes request task information of several IoT devices; A service cache module, used to input the state space information of the current time slot into the service cache model to obtain the action space information of the current time slot, wherein the action space information includes the action information corresponding to the tasks requested by several IoT devices; The task unloading decision module is used to obtain the task unloading decision information of the current time slot according to the state space information, data transmission information and action space information of the current time slot, and perform task unloading and execution according to the task unloading decision information of the current time slot.

9. A computer device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor; when the computer program is executed by the processor, the steps of the task offloading method based on the mobile cloud edge computing structure as described in any one of claims 1 to 7 are implemented.

10. A storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the steps of the task offloading method based on the mobile cloud edge computing structure as described in any one of claims 1 to 7.

Citation Information

Cited By

  • Dynamic cache management method and device of Internet of Vehicles and computer equipment

    CN122340178A

  • Dynamic cache management method, device and computer equipment for internet of vehicles

    CN122340178B