Research On Processing And Classification For Lidar Detection Signals

At present,global climate change has attracted people’s attention continuously,and atmospheric aerosols issue has become one of the hot topics of environmental research.Atmospheric aerosols is an important part of the earth-atmosphere system,it affects the balance of radiation between the earth and the atmosphere through direct and indirect radiation effects.As an active detection method,lidar has been widely used in atmospheric remote sensing detection with its high temporal and spatial resolution.Using lidar remote sensing data to retrieve the optical and microphysical parameters has become the main technique in the field of atmospheric remote sensing.The processing and analysis methods of lidar detection signals have a vital influence on the precise acquisition of aerosol optical parameters and microphysical parameters.This paper relying on the multi-wavelength lidar of the Lidar Atmospheric Remote Sensing Innovation Team Studio of North Minzu University as the experimental platform.Through the construction of optical path for lidar system and debugging of the data acquisition system,and then the established lidar system is used to conduct continuous experiments to collect,store and organize the data.In the process of lidar data acquisition,there are a lot of jump points in the collected data because of the strong atmospheric time variation and many interference sources.For this problem,the method of processing jump point of lidar detection data is proposed.The feasibility and reliability of this method are verified by using actual lidar data.The results show that the extinction coefficient profile can be successfully inverted when different remote boundary values are chosen by this method,and the extinction coefficient profile inverted is more continuous and smoother.It not only greatly increases the effective detection range of lidar,but also makes the inverted extinction coefficient profile more in line with the actual situation,which is more conducive to the subsequent further analysis of the atmospheric aerosol extinction coefficient profile.Based on the jump point processing,in order to improve the signal-to-noise ratio(SNR)of lidar detection,in view of the characteristics of lidar signal,a method of de-noising for lidar detection based on the improved variable weighted Kalman filtering algorithm is proposed.At the same time,this algorithm proposed in this paper is compared with classical Kalman filtering algorithm,error covariance weighted Kalman filtering algorithm and variable weighted Kalman filtering algorithm.The results show that the lidar detection signal can be de-noised effectively using this method,and the effective detection range of lidar can be increased.Compared with the other three Kalman filtering algorithms,the SNR of the lidar detection is improved 2.9d B,1.7d B and 0.6d B respectively under cloudy weather.In cloudless weather,the SNR of lidar detection is improved 2.5d B,1.4d B and 0.4d B respectively compared with the other three filtering methods.The improvement of lidar detection SNR is beneficial to the accurate acquisition of atmospheric extinction coefficient.It is important for the study of climate changes and their long-term trends.It also has important significance for the inversion of PM(Particulate Matter)value,determination of boundary layer,inversion of aerosol particle size distribution,and transport of air pollutants.Aiming at the classification problem of lidar detection signals,a classification method of lidar detection signal based on multi-scale entropy(MSE)is proposed,which aims to distinguish the complex characteristics of Lidar detection signals under different weather conditions by using the MSE method.The lidar detection signals of 1064 nm and 532 nm observed continuously in one day were preprocessed to obtain the range-squared correction signals,and they was calculated and analyzed under three different signal-to-noise ratios by MSE.By analyzing and comparing the MSE characteristics of lidar detected signals under three different types of weather,and combining with statistical methods to verify the effectiveness of the proposed method.The experimental results show that under the three sets of detection distances,MSE can effectively distinguish the haze degree of three different types of weather: severe pollution、moderate pollution and mild pollution.This paper validates the feasibility of the MSE method for analyzing and distinguishing lidar detection signal under three different types of weather,this method has the potential to become a new tool for distinguishing lidar detection signal under different types of weather,which has a great innovative significance for the processing and analysis of lidar detection signal.