Low-orbit satellite network congestion control method and device

The invention provides a low-orbit satellite network congestion control method and device, and the method comprises the steps: obtaining a network data training set in an offline state, and carrying out the reinforcement learning of a target agent. The training set comprises states and actions at di...

Full description

Saved in:
Bibliographic Details
Main Authors WANG ZIHAN, LIU YUANJIE, ZHANG JIAO
Format Patent
LanguageChinese
English
Published 05.12.2023
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:The invention provides a low-orbit satellite network congestion control method and device, and the method comprises the steps: obtaining a network data training set in an offline state, and carrying out the reinforcement learning of a target agent. The training set comprises states and actions at different moments, and time series data of states and reward values at the next moment. In the reinforcement learning process, based on a reward value, a target action value function and an estimated action value function, an expected quantile regression function is introduced to construct a TD loss function. And constructing a first loss function and a second loss function by using the state value function, the target action value function and the estimated action value function. And finally extracting a strategy function. And iteratively updating the parameters of all the functions. And in an online state, updating the network data training set based on time sequence data acquired by a real-time network. And utiliz
Bibliography:Application Number: CN202311089743