Multi-user access control method for UAV network based on deep reinforcement learning

Abstract

The invention belongs to the technical field of wireless communication, and relates to an unmanned aerial vehicle network multi-user access control method based on deep reinforcement learning. According to the invention, the deep reinforcement learning framework adapting to the multi-user access condition in the unmanned aerial vehicle network is provided by using the inherent change rule in the deep reinforcement learning environment, and the unmanned aerial vehicle network multi-user access control scheme based on the deep reinforcement learning under the condition that the global network information is unknown is realized. Compared with a traditional access control mode, the access control mode provided by the invention can realize higher system throughput and lower switching times. Andmeanwhile, different compromises can be realized by adjusting the switching penalty term under the conditions of throughput and switching times, and the performance can be guaranteed under differentswitching penalty conditions.

Description

technical field [0001] The invention belongs to the technical field of wireless communication, and relates to a multi-user access control method of an unmanned aerial vehicle network based on deep reinforcement learning. Background technique [0002] The traditional access control technology is realized by selecting different metrics (for example, received signal strength, etc.) and selecting an appropriate threshold value by means of threshold comparison. When the received signal strength of the user equipment (UE) from the source base station is lower than the set threshold, it will select a base station that can provide a received signal strength higher than the threshold for access. However, for the UAV network using UAVs as base stations, due to the mobility of the base station, the relative distance between the base station and the user changes frequently, resulting in drastic changes in the received signal strength at the user. At this time, the traditional access con...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problem of frequent switching of traditional access control technology in the UAV network and ensure the overall throughput of the network in the case of multi-user access, the present invention mainly focuses on the long-term throughput and switching times of the overall system

Due to the excellent performance of deep reinforcement learning in complex dynamic environment decision-making problems, in order to overcome the problem that global network information is difficult to collect in the UAV network environment, the present invention uses deep reinforcement learning to learn the inherent changes in the environment, and proposes an adaptive Using the deep reinforcement learning framework in the case of multi-user access in the UAV network, and realized the UAV network multi-user access control scheme based on deep reinforcement learning when the global network information is unknown

Images