A load balancing access method for Internet of Vehicles based on history reinforcement learning

Abstract

The present invention provides a load balancing access method of the Internet of Vehicles based on historical reinforcement learning. Firstly, the access base station allocation mode of the vehicle is obtained through the initial reinforcement learning module, and is continuously accumulated in the access mode accumulation library; after learning and accumulation, Let the history reinforcement learning module replace the initial reinforcement learning module and continue to run in the system. When the base station encounters network changes again, the history reinforcement learning module calls the access mode to accumulate historical records in the library, and learns new vehicles adaptively. The access allocation mode is recorded and an operating loop is formed to adaptively handle dynamic changes in the network, so as to ensure network load balance for vehicle access in a dynamically changing Internet of Vehicles environment. The present invention utilizes the potential regularity of the spatio-temporal distribution of the traffic flow, learns and utilizes the historical load-balanced vehicle access spatio-temporal experience from the iterative feedback with the environment, so as to continuously obtain the vehicle access base station allocation that can ensure the network load balance in the dynamic environment plan.

Description

technical field [0001] The invention relates to the technical field of network load balancing of the Internet of Vehicles, in particular to a load balancing access method of the Internet of Vehicles based on history reinforcement learning. Background technique [0002] With the development of the Internet of Vehicles, more and more vehicles need to access heterogeneous base stations in the network. These heterogeneous base stations are different in transmission power, physical size and construction cost. In a city, these access requirements are very different. For example, in dense traffic areas, the demand for vehicle access to base stations is much greater than that in sparse traffic areas. Under the traditional maximum SINR (Signal to Interference plus Noise Ratio) scheme, base stations with stronger power can attract more vehicles to access. A strong base station receives a strong downlink signal, which causes the base station with higher power to be overloaded, while t...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is how to ensure the load balance of the network when the vehicle accesses the base station in the case of heterogeneous base stations, unknown and complex Internet of Vehicles environment, and high dynamic changes of vehicles

Images