论文部分内容阅读
介绍基于混沌理论的海杂波分析方法,依据加拿大McMaster大学IPIX雷达实测海杂波数据,按照TAKENS的嵌入定理,设计了混沌预测器.预测器采用径向基函数神经网络重建海杂波的混沌动力系统,即建立海杂波的精确预测模型.实验表明,基于径向基函数的混沌时间序列预测器精度完全能够实现单步预测功能.学习误差为0.0232,预测误差为0.0277.
The sea clutter analysis method based on chaos theory is introduced, and the chaos predictor is designed according to the measured sea clutter data of IPIX radar of McMaster University in Canada and the TAKENS embedding theorem. The predictor uses radial basis function neural network to reconstruct the sea clutter chaos Dynamical system, that is to say, to establish an accurate prediction model of sea clutter.Experiments show that the accuracy of the chaotic time series predictor based on radial basis function can fully realize the single-step prediction function with a learning error of 0.0232 and a prediction error of 0.0277.