基于PCA和相位差校正法的风切变预警算法

为了提高相干激光雷达风切变预警算法的预警率，提出一种基于主成分分析（principal component analysis，PCA）和相位差校正法的风切变预警算法。首先，利用PCA对雷达回波信号进行低阶主成分向量提取以实现信号重构，提高信噪比。其次，采用基于快速傅里叶变换的相位差校正法对回波信号峰值频率进行校正，降低回波信号能量泄露。然后，分别利用旋转电机实验数据和小型飞机实验数据对所提算法进行性能评估。结果表明，采用所提算法估计相干激光雷达回波信号峰值频率的平均绝对误差为0.26 MHz，对实际低空风切变的预警率为92.31%。所提算法可以有效降低相干激光雷达回波信号频率估计误差，实现低空风切变预警。

Wind shear warning algorithm based on PCA and phase difference correction

In order to improve the early warning rate of coherent lidar wind shear warning algorithm, a wind shear warning algorithm based on principal component analysis (PCA) and the phase difference correction method is proposed. Firstly, PCA is used to extract the low-order principal component vectors of radar echo signals in order to reconstruct the signal and improve the signal-to-noise ratio. Then, the peak frequency of the echo signal is corrected by the phase difference correction method based on fast Fourier transform to reduce the energy leakage of the echo signal. The performance of the algorithm is evaluated by using the experimental data of rotary motors and small aircraft respectively. The results show that the average absolute error of estimating the peak frequency of the coherent lidar echo signal is 0.26 MHz, and the early warning rate of actual low-altitude wind shear is 92.31%. This algorithm can effectively reduce the frequency estimation error of the coherent lidar echo signal and realize the early warning of low-altitude wind shear.