谱估计理论在弹道数据参数化建模中的应用

针对高超声速飞行器防御作战中目标运动规律的描述问题，提出一种基于经典和现代谱估计理论的弹道数据参数化分析和建模方法。该方法将弹道数据视作非平稳随机信号，研究其变化规律：首先，通过线性消势法消除信号的线性趋势，将其转变为平稳信号，以便于进行谱估计；然后，采用Welch算法给出大致的谱图，结合该谱图和改进的协方差法确定自回归参数模型的阶数，以避免模型阶数选择准则引起的不确定性问题；最终，给出弹道数据的参数化模型。仿真结果表明，使用本文算法建立的弹道数据参数化模型与动力学模型具有较高的一 致性。

Application of Spectrum Estimation Theory in Parametric Modeling of Ballistic Data

A parametric analyzing and modeling method on ballistic data is proposed for describing the target motion in hypersonic vehicle defense based on the classical and modern spectrum estimation theory. In this paper, the ballistic data is studied as the non-stationary random signal and transformed into the stationary state via the linear detrend method to facilitate the spectrum estimation. Then the Welch algorithm is adopted to generate an approximate spectrum, with which the improved covariance method is employed to determine the order of the auto-regressive parametric model to avoid the uncertainty due to the model order selection criterion. Finally, the parametric model of ballistic data is established. The simulation results demonstrate that the original dynamics model and the parametric model established by the proposed algorithm share a high consistency.