针对多普勒天气雷达时间分辨率和方位分辨率提高后,回波强度和径向速度精度无法满足1 dB 和1 m/ s 业务要求的问题,采用基于距离过采样技术和白化滤波算法的改进谱矩估计方法,使估计精度大大提高。该方法通过在距离向提高采样频率,获取更多的回波信号样本数量,然后采用白化滤波算法去除回波信号距离向的相关性,提高回波信号的独立样本数;并以白化滤波后的回波信号为基础,改进现有脉冲对谱矩估计算法,提高谱矩估计精度。仿真和雷达试验结果表明,在回波信号信噪比较大时,谱矩估计改进算法比现有脉冲对算法的估计精度更高,且可解决多普勒天气雷达因时间分辨率和方位分辨率提高后引起的精度降低问题。

An improved method of spectral moment estimation based on range oversampling and whitening filter is developed to aim at the data-quality problems of echo intensity and velocity, which the operation accuracy need of 1 dB and 1 m/ s could not be met under the enhancement of Doppler weather radar azimuthal and time resolution. In this method, the range sampling-frequency is adopted to increase to acquire more numbers of radar echo signal, and the whitening filter is used to remove the range sampling-time correlation and increase the equivalent number of independent samples. Based on the whitened range oversampling echo signal, the algorithm of pulse pair processing is improved to enhance the accuracy of estimation. Simulation and experimental results indicate that the accuracy decrease can be remedied through these methods during the processing of azimuthal and time resolution under higher signal-to-noise and the accuracy need of radar operation application can be met.

TN957.71