A method and device for transmitting power distribution of UAV network based on reinforcement learning

Abstract

The invention discloses a method and device for distributing the transmission power of an unmanned aerial vehicle network based on reinforcement learning. The method includes: initializing the state of the unmanned aerial vehicle and the user, defining the action space and the parameters of the neural network; initializing the value function and the current network The Q value of the state; interact with the environment to receive reward feedback and calculate the Q value of the current action; select the optimal action according to the maximum value of the reward obtained from the feedback; update the next state of the system and store it in the experience pool; judge whether the amount of experience is Reach the expected value, if it is reached, start deep reinforcement learning, if not, continue the cycle; get instant reward according to Bellman equation and use DPPO algorithm for deep learning, and carry out reinforcement learning with the goal of maximizing long-term reward function; judge long-term reward Whether the value of the function tends to converge, the learning is terminated when it tends to converge, and the resource allocation optimization of the UAV wireless ad hoc network has been completed at this time.

Description

technical field [0001] The invention relates to the technical field of UAV-based wireless communication networks, in particular to a method and device for transmitting power distribution of UAV networks based on reinforcement learning in a multi-UAV-oriented wireless ad hoc network network scenario. Background technique [0002] Unmanned aerial vehicle (UAV) has become an important partner in the field of wireless communication technology to provide effective wireless connection services due to its flexibility and controllability, high mobility, and good air-to-ground line-of-sight links. , massive and other communication scenarios provide efficient solutions. UAVs can serve as mobile base stations in the air, providing high-quality communications for users on the ground and increasing the capacity of wireless networks. Compared with the traditional terrestrial communication system, flexibility and low time cost are the advantages of the UAV system. With UAV, LOS communica...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a UAV network transmission power allocation method and device based on reinforcement learning, to solve the problems of wasting channel resources and low power allocation efficiency in the resource allocation mechanism of the current UAV wireless ad hoc network. Artificial intelligence-based UAV wireless network resource allocation scheme

Images