最大概似法在多普勒激光测风雷达中的应用研究

Performance of maximum likelihood estimators of Doppler frequency for coherent Doppler lidar

为了从相干多普勒测风激光雷达的微弱气溶胶后向散射信号中高效、准确地提取多普勒频移信息,得到最接近CRB极限的无偏估计,利用气溶胶后向散射信号统计特性,建立了相干多普勒激光雷达仿真模型,并用最大概似法(ML)推导了在功率谱宽度已知情况下多普勒频率的估计算法。利用1.54μm全光纤相干激光雷达系统获得风场测量数据,用ML算法进行多普勒频移估计,并将结果与传统的周期图最大值法(PM)所得结果进行比较。对比表明,ML算法与PM算法的数据处理结果有较好的一致性,并且ML算法的多普勒频率探测能力更优。

To obtain the nearest estimate of the theoretical limit of the cramer-rao bound（CRB）,an estimator algorithm based on the maximum likelihood（ML） estimates is established,assuming known signal spectral width.The data of wind velocity was acquired by a 1.54 μm all fiber coherent Doppler lidar system,which was applied in the ML algorithm to estimate the Doppler frequency.The comparison of ML estimator and PM estimator shows that the ML algorithm has a better performance in Doppler frequency estimating.