In order to realize the accurate inversion of lidar data, the calibration method for the ratio of Fabry-Perot etalon (FPE) transmission to reflection curve is given, and the reliability of the method is verified by simulation test. The effective ratio of FPE transmission to reflection for the actual atmospheric detection is analyzed, and a nonlinear iterative algorithm for simultaneous inversion of wind field and backscatter ratio, as well as the specific iterative steps, initial value determination methods and corresponding error formula of parameters are proposed. The simulation results show that no matter whether the backscatter ratio R-beta is small or large, the iterative process is always convergent and the correct inversion results can be obtained. The value of R-beta has a great influence on the convergence rate of its own, but has little effect on the wind speed convergence rate. The smaller R-beta is, the faster convergence rate of R-beta is, vice versa. Error analysis inferred that under the shot-noise limit, the radial wind speed measurement error decreases rapidly with the increase of R-beta for 1 2; if the total number of backscattering photons of 50,000 received by telescope is assumed, the backscatter ratio measurement error increases with the increase of R-beta; within the wind speed measurement dynamic range of +/- 25 m/s, the radial wind speed error is below 2m/s and the backscatter ratio relative error is less than 4.1% when 1.1 < R-beta < 10. (C) 2018 Elsevier GmbH. All rights reserved.