火力与指挥控制2025，Vol.50Issue(9)：136-141,149,7.DOI:10.3969/j.issn.1002-0640.2025.09.017
基于循环神经网络的数据链优先级阈值预测方法
A Priority Threshold Prediction Method for Data Links Based on Recurrent Neural Networks
赵志勇 1毛忠阳 1徐建武 1潘耀宗1
作者信息
- 1. 海军航空大学航空作战勤务学院,山东 烟台 264001
- 折叠
摘要
Abstract
The rational and accurate setting of priority thresholds is crucial for enhancing the operational efficiency of data link networks.To resolve the shortcomings of existing priority threshold setting methods,a predictive method for data link priority thresholds based on recurrent neural networks(RNNs)is proposed,which integrates the advantages of artificial intelligence prediction algorithms with priority threshold setting in data link networks.In this method,the theoretical basis for dynamically setting priority thresholds is first derived to establish a link between priority thresholds and dynamic changes in data link network parameters.Subsequently,through the micro-slot processing of statistical time windows,training time series for the RNN are generated from observed data.A priority threshold prediction model is then constructed using gated recurrent unit(GRU)cells to achieve the prediction of priority thresholds.The real state of the network can be reflected accurately.The accuracy of data packet transmission decisions made by nodes within the network is improved.关键词
优先级阈值/循环神经网络/预测/数据链Key words
priority threshold/recurrent neural network/prediction/data link分类
信息技术与安全科学引用本文复制引用
赵志勇,毛忠阳,徐建武,潘耀宗..基于循环神经网络的数据链优先级阈值预测方法[J].火力与指挥控制,2025,50(9):136-141,149,7.
