9.7
NOAA advanced signal processing system for radar wind profilers
Daniel E. Wolfe, NOAA/ETL, Boulder, CO; and B. L. Weber, T. L. Wilfong, D. C. Welsh, D. B. Wuertz, and D. A. Merritt
The System Demonstration and Integration Division of NOAA's Environmental Technology Laboratory has developed an advanced Signal Processing System (SPS) that can be integrated with existing radar wind profiliers (RWP). The SPS consists of several modules designed to run in real time either at the RWP site or at a remote location. The modular design permits flexible implementation with either of the commercial-off-the-shelf RWP dwell engines (POP or LAPXM) as well as easy enhancement. An in-band filter (IBF) module can be applied to unaveraged spectra producing a number of statistics on each Doppler bin, including a mean Doppler spectrum. Other filters may also be applied to the spectral data including the standard averaging techniques. A signal detection (SigDetect) module calculates a noise and noise threshold level which are then used to determine whether or not a signal is present in each bin. A multiple moments (MultiMom) module then calculates the moments for each identified signal in each spectrum. A signal identification (SigID) module takes these data and determines which of the multiple signals best describes the atmosphere. The physical properties of each signal are examined along with three characteristics: persistence of the signal over range, over time, and across antenna beams. Finally, the meteorological product (MetProd) module produces wind products (speed, direction) for a time and height scale determined by the user. Results are presented in this paper from implementations of SPS on several RWP's, including a new 915-MHz multi-element electronically-stabilized antenna system, and a network class 449-MHz RWP and a 50-MHz RWP at Vandenberg Air Force Base, California. Comparisons with both balloon-derived data and conventional processing using consensus averaging show that SPS more effectively rejects interfering signals and can produce quality products at a much higher temporal resolution than conventional techniques.
Session 9, Radar Wind Profilers
Thursday, 18 January 2001, 8:00 AM-10:00 AM
Previous paper Next paper