摘要
针对相控阵雷达因跟踪资源限制导致的雷达散射截面积(RCS)序列采样时间非均匀问题,提出了一种基于高斯过程优化Lomb-Scargle算法的周期估计方法.该方法首先采用Lomb-Scargle算法处理非均匀采样序列,获得鲁棒的周期初值;然后构建包含周期核的高斯过程模型,以该初值为起点,通过对数似然函数对周期等超参数进行迭代优化.仿真结果表明,在低观测时长占比和低信噪比条件下,本文方法的均方根误差优于传统的三角函数拟合方法,显著提升了非均匀采样RCS序列的周期估计精度.
Aiming at the problem of non-uniform sampling time of radar cross section(RCS)sequence caused by the limitation of tracking resources in phased array radar,this paper proposes a period estimation method using a Lomb-Scargle algorithm optimized by Gaussian process.In this method,the Lomb-Scargle algorithm is first adopted to process the non-uniform sampling sequence and obtain robust periodic initial value;then,a Gaussian process model including the periodic kernel is constructed,and finally,this initial value is taken as the starting point to iteratively optimize the period and other hyperparameters through the log-likelihood function.Simulation results demonstrate that,under conditions of a low proportion of observation duration and a low signal-to-noise ratio,the proposed method achieves fewer root mean square error than traditional trigonometric function fitting,significantly improving the period estimation accuracy of non-uniformly sampled RCS sequences.
作者
秦晓东
白小二
韩文俊
俞建国
QIN Xiaodong;BAI Xiaoer;HAN Wenjun;YU Jianguo(Nanjing Research Institute of Electronic Technology,Nanjing 210039,China)
出处
《空天预警研究学报》
2026年第2期91-94,共4页
JOURNAL OF AIR & SPACE EARLY WARNING RESEARCH
