DQN-based multi-service low-orbit satellite resource allocation method

Abstract

The invention relates to a DQN-based multi-service low-orbit satellite resource allocation method, which belongs to the field of satellite communication, and comprises the following steps: S1, establishing a joint power and channel allocation model based on multiple services of a low-orbit satellite; S2, mapping resource allocation of a multi-beam low-orbit satellite communication system into a problem of obtaining maximum long-term income by interactive learning of an intelligent agent in an environment; and S3, solving the problem in S2 through state reconstruction and a DQN algorithm. According to the method, the system throughput can be improved under the conditions of meeting the multi-service user requirements and maintaining a service queue to be stable.

Description

technical field [0001] The invention belongs to the field of satellite communication, and relates to a DQN-based multi-service low-orbit satellite resource allocation method. Background technique [0002] As a supplement to the ground communication system, the low-orbit satellite communication system has irreplaceable advantages such as lower propagation delay and higher throughput, and is regarded as an important part of 5G communication. Due to the inherent laws of society and economy, different regions require different satellite services, resulting in uneven traffic distribution among beams. When the performance of the network demand is different from the preset capacity, it is difficult for the configuration of the satellite to adapt to this change, and the network is prone to congestion. Due to the limited bandwidth of the satellite, and the call will not end immediately, when the traffic increases, the impact of the current resource allocation results on the future e...

AI Technical Summary

Problems solved by technology

[0004] 1) GEO satellites are stationary relative to the ground, while low-orbit satellites move along the orbit at high speed, covering an area within about 5 to 12 minutes, so it is difficult to directly apply resource allocation algorithms to low-orbit satellite networks

[0005] 2) The resource allocation algorithm of most satellites still adopts the traditional iterative algorithm, and the artificial participation factor is obvious

[0006] 3) Some studies consider applying reinforcement learning technology to satellite system performance optimization, but still aiming at the optimization of a single resource

Images