Research For Visibility Detection By Small Lidar And Inversion Algorithm

Visibility reflects optical state of the atmosphere and indicates relevant changes. It plays an important role not only in meteorological observation, but in aviation, navigation, and transport and military affairs as well. Lidar has become the effective method to detect visibility and other optical parameters, owing to its short wavelength,high spatial and temporal resolution and detection sensitivity.The aim is measuring atmospheric visibility by use of the small self-developing Lidar. The basic structure of Lidar is designed based on the theory of Mie scattering. The selection standard of key components, processes of assembling and debugging the receiving signal are are discussed in detail. The overlap factor,key parameters and the effect to receiving signals by noise are analysed.The key of visibility measured by Lidar is the accurate inversion of extinction coefficient, so the inversion algorithms of extinction coefficient are well researched in the paper. Firstly, on the basis of Klett inversion algorithm, two improved idears are proposed. One is automatically estimating extinction efficient boundary values according to return signals and the SNR data. The other one is fitting far-field logarithmic range-adjusted power curve by least square fitting to reduce the higher error in far-field inversion results. Secondly, an inversion method taking advantage of Extended Kalman Filter is discussed in detail to overcome the defect that horizontal visibility is hard to detect precisely because of multi-scatter of atmosphere. Compared with other methods, improved algorithms mentioned above show the effectiveness on inversion accuracy of the extinction coefficient and the visibility by their Monte-Carlo simulation results. Fernald method is used to inverse the slant visibility.Finally, some improved algorithms are programmed by C# software and the horizontal and slant visibility results measured in our college are given, which prove the feasibility and reasonableness of the Lidar design and numerical simulation of the improved algorithms.