Software-defined space-air-ground integrated network routing optimization method based on deep reinforcement learning

Abstract

The invention discloses a software-defined space-air-ground integrated network routing optimization method based on deep reinforcement learning, and the method comprises the steps: firstly building a software-defined space-air-ground integrated network based on the thought of a software-defined network, and starting a controller to monitor and collect the state data of each node and link in the network; and establishing a deep reinforcement learning model according to network characteristics, taking the collected state data as the input of the deep reinforcement learning model, and outputting a link weight matrix of the network through training. When data forwarding is carried out, k paths are calculated by using a K-shortest path algorithm (KSP) and an alternative path set is formed, meanwhile, an appropriate path is selected to carry out data forwarding according to real-time monitoring of a controller on a link state, and finally, the convergence speed of a deep reinforcement learning model is improved by calculating an award value in a current state. Therefore, the optimization of the software-defined space-air-ground integrated network route is realized. The method not only can effectively adapt to the dynamically changing network topology, but also has the advantages that the average end-to-end delay and throughput are obviously improved compared with the existing method, and the data transmission efficiency of the space-air-ground integrated network is improved.

Description

technical field [0001] The present invention relates to the field of wireless communication, specifically, the present invention relates to a software-defined air-space-ground integrated network routing optimization method based on deep reinforcement learning. Background technique [0002] With the development of communication technology and the growth of Internet service requirements, users have continuously increased requirements for network communication range and network communication quality. Traditional ground-based networks have good communication quality, but cannot cover areas with harsh environments such as forests, mountains, and oceans. Space-based networks use satellites as relay nodes to ensure global coverage of signals. However, due to the impact of the space environment, space-based networks have problems such as long delays and high bit error rates. With the continuous improvement of user needs, the air-space-ground integrated network that combines ground-...

AI Technical Summary

Problems solved by technology

However, due to dynamic changes in network topology and poor link quality, the air-space-ground integrated network needs to establish an effective routing optimization strategy to improve network performance.

[0003] Due to complex characteristics such as topology dynamic changes, high bit error rate, and large transmission delay in the air-space-ground integrated network, it is difficult for the network to build a stable end-to-end transmission path on the basis of guaranteeing business service quality.

Due to the inability to respond to dynamically changing topologies in real time, traditional static topology routing algorithms cannot adjust corresponding routing strategies according to real-time changes in node and link states.

However, the dynamic topology routing algorithm has high requirements on the hardware conditions of the network and occupies a large amount of node resources, so it cannot fully adapt to the characteristics of limited node resources in the air-space-ground integrated network.

Images