LEO satellite channel allocation method based on reinforcement learning

Abstract

The invention discloses an LEO satellite channel resource allocation method based on reinforcement learning. The LEO satellite channel resource allocation method comprises the following steps: (1) a satellite centralized resource pool pre-allocates channel resources to each beam cell; (2) if the pre-allocated resources are surplus, the pre-allocated resources are collected to a resource pool, andif the pre-allocated resources cannot meet user requirements, channel resources are scheduled in a dynamic allocation mode; (3) the system performs training by using a Q-Learning algorithm to search for an optimal allocation strategy, and performs dynamic channel scheduling according to the allocation strategy after the training is finished; and (4) the system enters the next service request timeinterval, and allocates the channels in a mode of combining fixed channel pre-allocation and dynamic channel scheduling. According to the method, the channel resources are managed through the centralized resource pool to adapt to the inter-beam service difference, and the problem of time correlation in channel allocation is solved by adopting reinforcement learning, so that efficient channel resource allocation of the LEO satellite system is realized.

Description

technical field [0001] The invention relates to the technical field of LEO satellite application, in particular to a reinforcement learning-based LEO satellite channel allocation method. Background technique [0002] Among various types of satellites, low-orbit satellites have the characteristics of small path loss, short communication delay, and flexible orbital positions. Through low-orbit satellite constellations, seamless coverage of global regions can be achieved. However, resources such as spectrum, power, and time slots available for satellite systems are extremely scarce and valuable, and it is an urgent problem to allocate satellite network resources reasonably and efficiently. [0003] Due to the dynamic changes in the coverage area caused by satellite movement and the non-uniform distribution of ground users, the traffic load is changing all the time, and the on-board resources of low-orbit satellites have been solidified when they leave the factory. The tradition...

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a LEO satellite channel allocation method based on reinforcement learning, which can realize cross-beam scheduling of LEO satellite channel resources, thereby adapting to business differences between beams, and uses Q-Learning algorithm to learn optimal channel allocation strategy and perform dynamic channel allocation. Allocation, to solve the time correlation problem in the system channel allocation process

Images