Secure relay node selection method based on reinforcement learning in fog computing

Abstract

Along with continuous development of fog calculation, more safety communication problems gradually emerge from the water surface while the information processing speed is increased. In order to improve the key generation rate, the invention provides an optimal double-relay node selection method based on Q-learning in order to solve the problem of how to select a safe optimal double-relay node. Firstly, a safety fog calculation structure model based on social awareness is constructed; then, an optimal double-relay node selection method based on a Q-learning algorithm is designed under the model; the selection of the optimal double-relay node in a dynamic environment is realized; finally, the key generation rate, the selection speed of the double relay nodes and the selection accuracy of thedouble relay nodes in the dynamic environment are analyzed, the optimal double relay nodes can be effectively selected in the dynamic environment, the algorithm is quickly converged to be stable, andthe selection speed of the optimal relay nodes is effectively increased.

Description

technical field [0001] The present invention selects the relay node by using the characteristics of the wireless signal channel, and uses the reinforcement learning Q-learning algorithm to select the optimal relay node, so as to realize the improvement of the key generation rate of the physical layer. Physical layer key generation belongs to the field of communication, and using reinforcement learning to select relay nodes belongs to the field of computing. Background technique [0002] In the past few years, the explosive growth of mobile network traffic reflects the public's increasing demand for mobile communications. At the same time, a variety of mobile applications have emerged on the market. These applications not only increase the computing burden of user equipment, but also increase the user's requirements for fast real-time communication. Although cloud computing achieves the former, it cannot meet the low-latency network requirements due to its geographical locat...

AI Technical Summary

Problems solved by technology

At present, in the research of applying traditional security technology to fog computing network, some researchers try to apply security solutions in cloud environment, such as cloud access control, key management, digital signature, identity authentication, etc. to fog computing environment. , but due to the multi-layer structure of the mobile fog computing network and the mobility of terminals among fog nodes, the online distribution, maintenance and management of mobile terminal keys become extremely difficult

Images