基于IAVOA-GRU网络的高频地波雷达电离层杂波预测

Ionospheric Clutter Prediction of High Frequency Surface Wave Radar Based on IAVOA-GRU Network

电离层杂波的精确预测对提升高频地波雷达的探测性能具有重要作用。提出了一种基于改进非洲秃鹫优化算法优化门控循环单元(Improved African Vultures Optimization Algorithm Optimization Gated Recurrent Unit,IAVOA-GRU)网络的电离层杂波预测方法。首先,依据电离层杂波的混沌特性,通过相空间重构方法对接收到的电离层杂波进行相空间重建,构建GRU网络的输入、输出样本集;然后,利用IAVOA对GRU网络的隐层节点数、迭代次数及初始学习速率3个超参数值执行优选;最后,重新训练优化后的GRU网络,并进行预测。实测结果表明,相较其他6种对比预测模型,所提出的IAVOA-GRU网络模型具有较高的预测精度和可靠性,为有效改善高频地波雷达的探测性能提供了一种思路和方法。

Accurate prediction of ionospheric clutter plays an important role in improving the detection performance of high frequency surface wave radar(HFSWR).An ionospheric clutter prediction method based on Improved African Vultures Optimization Algorithm Optimization Gated Recurrent Unit(IAVOA-GRU)network is proposed.Firstly,according to the chaotic characteristics of ionospheric clutter,the phase space reconstruction method is used to reconstruct the received ionospheric clutter,and the input and output sample sets of GRU network are constructed.Secondly,IAVOA is used to optimize the number of hidden layer nodes,iteration times and initial learning rate of GRU network.Finally,the optimized GRU network is retrained and predicted.The measured results show that the proposed IAVOA-GRU network model has higher prediction accuracy and reliability than other six prediction models,which provides an idea and method for effectively improving the detection performance of HFSWR.