Signal Processing Of Radar Wind Profiling Based On Time-frequency Analysis

The wind profiling radar is a remote sensing equipment which is used to detect the wind field and other physical quantities of the atmospheric by atmospheric scatting of electromagnetic waves. Its detection objects is clear sky atmosphere, and can provide information of the meteorological elements of the atmospheric wind field, turbulence flow field, wind shear condition with high spatial and temporal resolution and high accurary. Also, it has been widely used in atmospheric exploration, aerospace and other fields. Now, the hardware technology on wind profiling radar of our country is advanced in the world, but it falls behind the abroad in signal analysis, spectrum moment processing and quality control. Therefore, it’s of great significance to devolop the signal processing algorithm and application research on wind profiling radar.First, we analyze the present domestic and foreign research on wind profiling radar and summarize the signal characteristics of the typical interference sources for wind profiling radar like ground clutter, intermittent clutter. Next, we introduce the classic signal processing method of wind profiling radar based on fast Fourier transformation (FFT), rank several kinds of the improved algorithms and point out corresponding shortage in the application.The method in term of time-frequency analysis has been widely used for processing the non-stationary signal. In time-frequency analysis, Fractional Fourier transform (FRFT) could provide the Chirp signal with the characteristics of energy clustering in the fractional Fourier domain. Motivated by this idea, this paper introduces a clutter suppression method based on FRFT. Besides, we also supply the deduice process and realization way of detailed algorithm. The algorithm is verified through two representative simulation examples, i.e. the case that single or double intermittent clutters exist in the wind profiling radar echo signal respectively. Through the comparison between our algorithm and the classic algorithm, our algorithm is proved to be effective in the clutter suppression. At last, the signal-to-clutter improvement factor is used to analyze the performance of the algorithm, and signal-to-clutter ratio obtained after clutter suppression has been improved greatly than that before clutter suppression. At the same time, it can be found that there is approximately linear relation between the signal-to-clutter ratio before the clutter suppression and that after the clutter suppression.After confirming that FRFT could suppress the intermittent clutter, for the large calculated quantity on the clutter detection by FRFT, I introduce a clutter suppression method based on the combination of Zak transform and FRFT by the use of the time-frequency distribution characteristics of the Zak transformation of intermittent clutter. The detailed realization of algorithm is given in this paper.Through the comparison between this algorithm and the classic algorithm, it can be found that the effect of clutter suppression is clear. Compared with FRFT scan method, this method greatly reduces the computational complexity of the Chirp detection by employing one-dimensional search instead of two-dimensional.This paper focuses on the suppression of intermittent clutter in the wind profiling radar signals. However, the actual wind profiling radar data will be affected by a variety of source of interference, any kind of single method is impossible to remove all interference signals. It will be a key job to combine the method in this paper with other methods for the remove of clutter in the wind profiling radar signals in the future.