基于周期图最大似然算法的相干激光测风多普勒频率估计

Doppler Frequency Estimation of Coherent Doppler Wind Lidar Based on Periodogram Maximum Likelihood Algorithm

采用径向风速的估计不确定度和探测概率作为评价指标,研究了周期图最大似然(PML)算法的多普勒频率估计性能.基于大气分层激光回波模型,分别以PML和周期图最大值法(PM)对回波信号进行处理,验证了PML算法在相干激光测风中的可行性;分析比较不同信噪比条件下PML算法的风速估计不确定度与探测概率.仿真结果表明,在发射脉冲宽度为400ns、采样点数为128时,PML算法适合在中等信噪比条件下使用,且风速估计的不确定度整体小于PM算法的,在信噪比为-13dB时径向风速的估计不确定度为0.75m/s,探测概率在90%以上,该研究为后续的外场试验提供了指导.

Using the estimation uncertainty of the evaluation index, the Doppler frequency estimation radial velocity and the detection probability as the performance of Periodogram Maximum Likelihood （PML） algorithm was studied. Based on the stratified atmosphere laser echo model, the echo signal was processed by PML and Periodogram Maximum method （PM） respectively to verify the feasibility of PML algorithm. The uncertainties of estimated wind speed and detection probability were analyzed comparatively under the different signal-to-noise ratios condition with PML algorithm. The simulation results show that, with a transmitting pulse width of 400ns and the sampling points of 128, the PML algorithm is suitable for moderate signal-to-noise ratio, and the overall uncertainty of the estimated wind speed is lesser than PM algorithm. When the signal-to-noise ratio is -13 dB, the radial velocity estimation uncertainty is 0. 75 m/s and detection probability is more than 90%. This study provides guidance for subsequent field trials.