基于谱聚集度的旋翼微动特征提取方法

Micro-motion feature extraction method based on concentration of spectrum

针对无闪烁条件下多分量周期性微多普勒信号的参数估计问题,提出了一种基于谱聚集度指标的微动目标特征提取方法.首先通过散射点模型分析了目标的微多普勒调制特性,然后利用目标回波慢时间维的相位信息构建微多普勒信号的解调算子,并引入奇异值比谱法估计微多普勒信号中的主周期分量,实现搜索空间的降维优化处理,最后通过计算微多普勒信号与解调算子乘积的谱聚集度来实现目标叶片数、叶片长度和初相的参数估计.仿真结果表明:该方法估计精度高、运算速度快,且在低信噪比和搜索网格失配条件下具有较好的鲁棒性.

Aiming at the parameter estimation of multi-component micro-Doppler signal without flicker,a micro-motion feature extraction method of target based on the concentration of spectral was proposed.Firstly,the micro-Doppler modulation characteristics of the target were analyzed by the scattering point model.Then,the demodulation operator of the micro-Doppler signal was constructed by using the phase information of the slow time dimension of the target echo,and the singular value ratio spectrum method was introduced to estimate the main period component of the micro-Doppler signal to realize the dimensionality reduction and optimization of the search space.Finally,the parameters of blade number,blade length and initial phase were estimated by calculating the spectral concentration of the product of micro-Doppler signal and demodulation operator.Simulation results show that this method has high estimation accuracy,fast operation speed,and good robustness under the conditions of low signal to noise ratio and search grid mismatch.