为了获得更为准确的大气风速廓线，使用国内首台基于三通道Fabry-Perot标准具的瑞利散射测风激光雷达进行风场测量，提出了根据实测数据及时调整标准具位置的激光频率反演方法。此方法有效减小了激光频率漂移和抖动造成的反演误差，同时对传统风速算法进行了改进。利用非线性迭代算法处理数据，通过对两种算法处理结果的比较，发现迭代算法具有更高的反演精度。使用非线性迭代算法对三通道瑞利散射多普勒测风激光雷达风场测量数据进行处理，得到了10km~40km的大气风速廓线。结果表明，此种风速反演算法切实可行，并提高了反演精度。

In order to get the more exact profile of horizontal wind speed, the wind field was measured with the first domestic Rayleigh Doppler wind lidar based on a triple Fabry-Perot etalon, a method was brought up to retrieve the laser frequency in which the location of the etalon was adjusted in time according to the real detected data. The errors caused by the jitter and drift of the laser can be effectively reduced by means of this wind inversion algorithm. In addition, improving the traditional method, a nonlinear iteration inversion method was proposed. After processing the same Monte-Carlo simulated data with both the methods, it was found that the nonlinear iteration inversion method had a better precision. The profile of horizontal wind speed from 10km to 40km was obtained after processing the Rayleigh Doppler wind lidar signal data by means of the nonlinear iteration inversion method. The results show that this inversion method is feasible, and can effectively improve the inversion precision.