基于小波分析和神经网络的卫星钟差预报性能分析

发布时间：2018-07-03 10:43

本文选题：天体力学 + 人造卫星　；参考：《天文学报》2010年04期

【摘要】：为了有效地进行卫星钟差预报和更好地反映卫星钟差特性,提出了一种基于小波分析和神经网络的4阶段混合模型来实现卫星钟差的预报,并给出了基于小波分析和径向基函数(Radial Basis Function,RBF)神经网络进行卫星钟差预报的基本思想、预报模型和实施步骤.采用"滑动窗"划分数据,利用神经网络预测小波分解和去噪后的钟差序列各层系数,更精确地把握钟差序列复杂细致的变化规律,从而更好地逼近钟差序列.为验证该混合预报模型的可行性和有效性,利用GPS卫星钟差数据进行钟差预报精度分析,并将其与灰色系统模型和神经网络模型进行比较分析.仿真结果显示,该模型具有较好的预报精度,可为实时GPS动态精密单点定位提供较高精度的卫星钟差.
[Abstract]:In order to effectively predict the satellite clock difference and better reflect the satellite clock difference characteristics, a four-stage hybrid model based on wavelet analysis and neural network is proposed to realize the satellite clock error prediction. The basic idea of satellite clock error prediction based on wavelet analysis and radial basis function (RBF) neural network is presented. The sliding window is used to divide the data and the neural network is used to predict the coefficients of every layer of the clock difference sequence after wavelet decomposition and denoising, so that the complex and meticulous variation law of the clock difference sequence can be grasped more accurately, so that the clock difference sequence can be approximated better. In order to verify the feasibility and effectiveness of the hybrid prediction model, the accuracy of clock error prediction is analyzed by using GPS satellite clock error data, and compared with the grey system model and neural network model. The simulation results show that the model has good prediction accuracy and can provide high precision satellite clock difference for real-time GPS dynamic precision single point positioning.
【作者单位】：电子科技大学电子科学技术研究院;
【基金】：航空科学基金(20090580013)资助
【分类号】：P127.1

【参考文献】