MEC-assisted intelligent resource management method in vehicle networking communication system

Abstract

The invention discloses an MEC-assisted intelligent resource management method in an Internet of Vehicles communication system, which adopts orthogonal frequency division multiplexing to convert frequency spectrum resources into mutually independent frequency spectrum sub-bands on different sub-carriers, and designs channel multiplexing of a plurality of V2I links and a frequency spectrum sub-band allocation strategy so as to minimize the time cost of the system. A multi-agent reinforcement learning algorithm is adopted, firstly, a state space, an action space and a reward function of a system are combined to model a problem into a Markov decision process, secondly, a multi-agent reinforcement learning algorithm path framework is established, and finally, training is performed to obtain an optimal system resource allocation strategy, so that resource management of the communication system is realized. And the overall time delay of the Internet of Vehicles system is reduced. Compared with a traditional algorithm, the algorithm can obtain a joint optimization solution, can be suitable for a rapidly changing channel environment, remarkably improves the information transmission rate between the vehicle and the base station and the system resource utilization rate, and reduces the task execution time delay of a communication system.

Description

technical field [0001] The invention relates to an intelligent resource management technology in a vehicle networking communication system based on MEC assistance, and belongs to the technical field of communication. Background technique [0002] With the development of society and the advancement of science and technology, the number of motor vehicles is growing explosively. The growing number of cars poses enormous challenges in terms of traffic, environment and human health. On the one hand, people have higher and higher requirements for driving safety, traffic efficiency, and driving experience; on the other hand, managers have increasing demands for automated traffic management, traffic control, congestion relief, and environmental protection. How to solve the above problems in the Internet of Vehicles has become the research content of the whole society. The development of cloud computing technology based on edge computing provides a more flexible and effective solut...

AI Technical Summary

Problems solved by technology

[0005] Sun W, E G, The SOLEN algorithm proposed by F et al. in the document "Radio resource management for D2D-based V2V communication" can only be applied to slowly changing channel fading, and cannot be applied to the rapidly changing channel environment in the Internet of Vehicles scenario and does not have the ability of adaptive learning

Images