| 摘要: |
| 大型低轨星座的部署使卫星数量迅速增长，随遇接入测控技术的应用使航天测控网能够完成更为密集的跟踪任务。在高密度跟踪任务中，针对现有测控设备集中监控面临的故障告警信息传递慢、业务监控与设备监控不联动等问题，提出了一种随遇接入业务驱动的测控设备全景监控系统。首先设计了由业务层、接口层和设备层组成的测控设备全景监控系统架构，其次在流量监测方面应用长短期记忆神经网络对随遇接入流量进行预测，并设计了3种核心监控指标辅助系统故障告警的排查。基于航天测控网数据的随遇接入流量预测实验表明，使用长短期记忆神经网络的流量预测值与测量值的均方根误差小于2，能够为全景监控系统的流量预测提供有效的告警门限。 |
| 关键词: 航天测控设备 随遇接入 全景监控 长短期记忆神经网络 随遇接入流量 |
| DOI：10.20079/j.issn.1001-893x.241121004 |
|
| 基金项目: |
|
| Design of a Random Access Driven Panoramic Monitoring System for Space TT&C Equipment |
| NING Peijie,ZHONG Dexing |
| (Faculty of Electronic and Information Engineering，Xi’an Jiaotong University，Xi’an 710049，China) |
| Abstract: |
| The deployment of low Earth orbit(LEO) mega constellations has led to a rapid increase in the number of satellites，and the application of random access TT&C technology enables the TT&C network to handle more intensive tracking tasks.In high-density tracking missions，existing centralized monitoring systems for TT&C equipment face challenges such as slow fault alarm information transmission and a lack of linkage between tracking monitoring and equipment monitoring.To address these issues，a panoramic monitoring system for TT&C equipment driven by random access operations is proposed.Firstly，a system architecture comprising a tracking layer，an interface layer，and an equipment layer is designed.Secondly，in terms of traffic monitoring，a long short-term memory(LSTM) neural network is applied to predict random access traffic，and three core indicators are designed to assist in troubleshooting system fault alarms.The experiment on traffic prediction based on data from the TT&C network demonstrates that the root mean squared error between the predicted and measured traffic values using the LSTM neural network is less than 2，which can provide an effective alarm threshold for traffic prediction in a panoramic monitoring system. |
| Key words: space TT&C equipment random access panoramic monitoring long short-term memory neural network random access traffic |
《电讯技术》编辑部 