Atmospheric radar signal processing using new algorithm and advanced techniques

Abstract

Radar probing of the atmosphere entered a new phase in early seventies through the pioneering work of Woodman and Guillen demonstrating that high power VHF radars offer a powerful means to explore the structure and dynamics of the middle atmosphere with unprecedented height and time resolutions. Their original work, employing the Jicamarca radar, led to the concept of Mesosphere- Stratosphere-Troposphere (MST) radar and this class of radars has come to dominate the scene of atmospheric probing over the past few decades. Nowadays atmospheric radars are extensively used for obtaining the wind information (wind profiler) by remote sensing of the atmosphere from ground. The Wind Profiling Radar System is an extremely versatile and high performance instrument for studies of the lower and middle atmosphere. Wind profilers data are widely used in the meteorological community for research and for the initialization of forecast models. However, any systematic velocity errors, such as those caused by any interference, make these data potentially less valuable or even damaging for analysis and prediction. Therefore identification of atmospheric signals from the radar back scattered echoes and computation of wind velocity is significant in the atmospheric radar data processing. It is possible with the sophisticated signal processing techniques to improve the accuracy of the wind velocities measured by the wind profilers. Continued integration of wind profiling technology into operations and research requires continued improvement in the reliability and accuracy in the derived meteorological products. In particular, extracting measurements of meteorological quantities in the presence of interfering signals and quantifying the error in the measurements introduced by non-homogeneous and other limiting meteorological conditions must be addressed. Certainly, improvements in profiler hardware offer some advantages and must be pursued; however, these improvements will be incremental. Significant improvements are possible through signal processing advances discussed in this thesis.