home » about nssl » research divisions » radar r&d » doppler radar & remote sensing
Signal processing
IDENTIFYING PROBLEMS AND FINDING SOLUTIONS:
Radar Improvements
PROBLEM: A need for faster, cleaner, better signal processing
SOLUTION: NSSL's polarimetric radar serves as a testbed where evolutionary and radical improvements in computing and signal processing are evaluated to determine if they are suitable to be included in the network of NWS radars.
SUCCESSES:
- Simultaneous transmission: Most dual-polarization radars alternate between the horizontal and vertical mode on a pulse-by-pulse basis through the use of high power, high-speed dual polarization switches. By changing from alternating to simultaneous dual polarization, the current issues of long dwell times and reduced velocity ranges are eliminated so scan time is faster. Simultaneous transmission also reduces costs because the expensive and difficult-to-maintain high power polarization switch is no longer needed.
- Whitening and oversampling techniques: NSSL developed a scheme to speed up the processing of weather radar data that involves sampling weather radar signals more frequently ("oversampling") to potentially increase severe weather warning lead-times. A mathematical transformation is then performed that decorrelates the data ("whitening") to solve a variety of signal processing problems. This method increases the speed of volume coverage by weather radar so that hazardous features can be detected faster, and leads to better estimates of precipitation and wind fields. The scheme works best on strong echoes. Weak echoes would still require standard filtering. This new development applies to any weather radar and has exciting implications for the weather forecasting and research community. Observations at very small intervals are required to understand the details of a vortex as it forms and dissipates near the ground. Even faster rates of volumetric data are required to determine the presence of transverse winds.
- Mitigation of range and velocity ambiguities: In most weather radars, the range and Doppler velocity ambiguity problems are coupled, such that trying to alleviate one of them worsens the other. Special techniques are necessary to resolve both ambiguities to the levels required for the efficient observation of severe weather. Efforts in this area are expected to culminate in superior radar data quality when implemented operationally. The increased data quality will result in an improved ability for the WSR-88D to detect severe weather, flash floods, winter storms, and provide aviation forecasts. (more» )
- Spectral processing. Spectral analysis of radar signals is being investigated to reduce data artifacts, measure wind fields by resolving contamination by biological scatterers, and retrieve information from weak signals that were previously discarded as unreliable. Besides using classical techniques of spectral analysis, NSSL devised a new approach for the computation of the polarimetric variables. Polarimetric spectral processing is used to determine the presence of multiple scatterer types in the resolution volume (e.g., water, insects, or birds), and to explore the intrinsic polarimetric properties of these scatterers. These advanced signal processing techniques make possible the identification, discrimination, and classification of radar echoes. In addition, spectral processing of polarimetric signals opens new possibilities for data censoring that could lead to improved retrieval of weak weather signals.
- Other areas of research include techniques for reducing the effects of artifacts and combined use of weather and profiling radars.