Research On Wind Profile Radar Data Processing Of Spatial Spectral Moments Estimation

With the rapid economic and social development, human traces on the weather became increasingly obvious that the emergence of extreme weather more and more. China is the country in the world that it is most types of natural disasters, most active, the most affected countries. Wind Profiler as a new generation of high-altitude atmospheric sounding system to detect clear air turbulence as the goal can be realized on remote sensing of atmospheric wind field, to provide real-time three-dimensional atmospheric wind field information, and increase Radio acoustic detection system (RASS), and microwave radiometer or GPS/MET with water vapor monitoring system can be realized in the atmosphere wind, temperature, humidity and other elements of continuous remote sensing. According to the plan, by 2020 China plans laid about 200 tropospheric wind profiling radar, and meteorological comprehensive probe into the business system to improve the general upper air sounding data in the relative lack of spatial and temporal density. In urgent applications, the paper-based spatial spectrum estimation (SSE) theory and algorithms, studied and implemented the use of spatial spectrum estimation of wind profile radar data processing process.According to the Doppler principle, wind profile radar echo to the frequency domain in order to extract the atmospheric wind field information to power spectrum estimation is wind profile radar data processing, a necessary step. This paper introduces the classic wind profile radar signal processing and data processing methods, and in-depth analysis of the establishment of signal processing based on I/Q data flow and the power spectrum after FFT, the data processing. On this basis, this paper focuses on the research and comparison of the more commonly used at home and abroad radar data processing algorithms, advantages and disadvantages. The study found that:in the wind profile radar spectral data processing, radar detection of low sample volume of ground clutter spectrum of data often occurs, intermittent clutter, clutter and the atmosphere back to the spectrum of geomagnetic mixed overlap situation. At present, the radar equipment manufacturers only on signal processing, and the classical estimation method is not efficient. But a serious shortage of data processing, it is difficult to obtain high-quality, high-sounding data consistency.For these reasons and application needs, the context of detection of different weather data in meteorological is analyzed. In the background of clear sky weather conditions, the proposed minimum variance spectral estimation (MVSE) algorithm, and gives the derivation of specific algorithm and implementation process, through simulation analysis shows that:the application of the algorithm significantly enhance and improve the the detection data processing efficiency and quality. Precipitation in the meteorological background conditions near the proposed use of maximum likelihood estimation and subspace fitting (MLE-SF) algorithm, experimental detection data network simulation the superiority of the MLE-SF, and compare the classical algorithm, proved that the algorithm has less computation, the implementation of high efficiency, for feasibility and other characteristics.Firstly, this article parameter model based on the minimum variance modern spectral estimation (MVSE) spectrum estimation algorithm to data in the clear sky background weather conditions, the use of simulation analysis of the actual probe data show that the algorithm can effectively suppress and remove clutter, and improve the wind profile radar detection range and accuracy, to the atmosphere back to the spectrum of the spectral moments of effective estimates. Secondly, the use of MLE-SF data spectral estimation algorithm is given a detailed derivation of the algorithm and the algorithm flow chart of the background near the precipitation weather conditions, the simulation results analysis, and classical data processing method were compared. Show that the algorithm is feasible, effective and practical, the sampling volume in the lower spectrum moment estimation has been improved significantly. Again, in the meteorological background of complex weather conditions, will MVSE, MLE-SF, FFT spectral estimation results comparing simulation results show that:MLE-SF precision than MVSE, compared with the classical FFT estimation method to obtain the spectral moments data with higher confidence in the sampling volume at the lower detection get better results, improve the accuracy of the spectral moment estimation. MLE is an important high-resolution spatial spectrum estimation method, MLE-SF estimation performance; in the case of coherent sources can still valid estimates. Finally, combining the work of this paper, summarized, and the prospect of future research work.