A method and device for controlling an intelligent reflective surface based on deep reinforcement learning

Abstract

The present invention provides a method and device for controlling an intelligent reflective surface based on deep reinforcement learning, wherein the method includes: a policy network generates a first action according to a first state; the amplitude is fixed and input into an optimization module, and the first action is updated to obtain a second action action, and get the first target value at the same time; apply the second action to the wireless environment to get the second state, get a new sample and store it in the experience pool; the policy network and the value network carry out deep deterministic policy gradient training according to the sample, and the executor uses The deep deterministic policy gradient method updates its parameters; determines the third target value according to the first target value and the second target value generated by the target Q network, trains the DNN of the online Q network according to the third target value and updates its parameters; repeats the above The steps are until the network parameters that minimize the transmit power of the AP are obtained and output. The invention can realize stable and high-efficiency learning in a shorter time, and can converge to the optimal goal faster.

Description

technical field [0001] The invention relates to the technical field of wireless communication networks, in particular to a control method and device for an intelligent reflective surface based on deep reinforcement learning. Background technique [0002] At present, IRS is considered to be a very potential and promising technology. IRS is composed of a large number of passive reflection elements, which are connected to each other and controlled by an embedded IRS controller. It can be used to improve the energy and spectrum of wireless communication. efficiency. Signal strength at the receiver is enhanced through joint control of the complex reflection coefficients of all reflective elements, known as passive beamforming. The passive beamforming technology of the IRS and the transmission control technology of the transceiver can further improve the network performance. IRS has been applied in various scenarios, which play different roles in wireless communication, such as ...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a control method and device for intelligent reflective surfaces based on deep reinforcement learning to solve the technical problems of low learning efficiency and poor stability of traditional deep reinforcement learning methods

Images