Self-adaptive load balancing ground user access method of unmanned aerial vehicle auxiliary network

According to the adaptive load balancing ground user access method of the unmanned aerial vehicle auxiliary network, based on an unmanned aerial vehicle deployment algorithm of a deep Q learning network (DQN) and adaptive and load balancing (ALB) for GUs access, the GUs access problem in a BS-UAV-NTN network is converted into a maximization problem, and the maximization problem is converted into a Markov decision process (MDP) problem of unmanned aerial vehicle deployment in an unknown environment. The method comprises a DQN-based unmanned aerial vehicle deployment algorithm and an access scheme for performing priority ranking on the BSs and the unmanned aerial vehicles. Simulation results show that the access scheme is superior to traditional Q-learning and random schemes in the aspects of rewards and the number of accessed GUs. 无人机辅助网络的自适应负载均衡地面用户接入方法，基于深度Q学习网络(DQN)的无人机部署算法和用于GUs访问的自适应和负载平衡(ALB)，将BS-UAV-NTN网络中的GUs接入问题化为一个最大化问题，将其转化为未知环境下无人机部署的马尔可夫决策过程(MDP)问题。该方法包括一种基于DQN的无人机部署算法，以及一种对BSs和无人机进行优先级排序的接入方案。仿真结