Edge computing task unloading method based on reinforcement learning

Abstract

The invention discloses an edge computing task unloading method based on reinforcement learning. The method comprises the following steps: S1, establishing an edge computing system model; s2, a common user node generates a task and reports the task to an SDN master controller node, local calculation or unloading calculation is selected according to the user task condition, and when unloading calculation needs to be carried out, the SDN master controller node trained through a reinforcement learning method carries out unloading calculation according to the real-time network state, the spectrum resources and the calculation resources; obtaining an optimal unloading strategy based on the network state of the edge node and feeding back the optimal unloading strategy to the common user node; s3, the common user node unloads the task to the service node according to the optimal unloading strategy, and the service node executes the distributed calculation task and feeds back a calculation result to the common user node. According to the invention, the problem of computing resource shortage can be effectively relieved.

Description

technical field [0001] The present invention relates to the technical field of wireless communication, and more specifically relates to an edge computing task offloading method based on reinforcement learning. Background technique [0002] With the advanced development of communication and computing technology, the number of devices with networking requirements is already very large, and the functions of devices are also diversified. For example, predictable Internet of Things applications such as Internet of Vehicles and smart medical care are also booming. The tolerable delay of tasks has very strict requirements, and the huge number of networked devices will inevitably lead to bursts of data traffic. These pose severe challenges to the centralized computing processing and large-scale data storage of traditional cloud computing. There are the following problems: 1) Delay: The device is far away from the cloud computing center, which will have a serious impact on some delay...

AI Technical Summary

Problems solved by technology

[0002] With the advanced development of communication and computing technology, the number of devices with networking requirements is already very large, and the functions of devices are also diversified. For example, predictable Internet of Things applications such as Internet of Vehicles and smart medical care are also booming. The tolerable delay of tasks has very strict requirements, and the huge number of networked devices will inevitably lead to bursts of data traffic. These pose severe challenges to the centralized computing processing and large-scale data storage of traditional cloud computing. There are the following problems: 1) Delay: The device is far away from the cloud computing center, which will have a serious impact on some delay-sensitive devices. For example, in the Internet of Vehicles, delay may pose a serious threat to human life; 2) Traffic: The exponential growth of the number of devices will inevitably lead to an increase in traffic, communication congestion, and reduced user experience quality; 3) Security and energy consumption: When devices transmit information, they need to go through a long path, which may lead to risks such as data loss or information leakage ; The high energy consumption caused by the high load of the data center is also the core problem of cloud computing

Images