WEATHER RADAR USING SPECTRAL GAUSSIAN ENVELOPE DISCRIMINATION FOR CLUTTER REJECTION

摘要

A method for removing clutter related frequency components from power spectrums generated from weather radar return signals to provide improved windshear detection capability. Weather radar return signals are synchronously detected and digitized to provide i and q time domain sample sequences. The i and q time domain sample sequences are passed through a window function and then transformed to frequency domain sequences by a Fast Fourier Transform. A power spectrum is generated from the frequency domain sequences. The spectrum is subjected to a Spectral Gaussian Envelope Discrimination (SGED) process in which the spectral envelope is scanned to identify any lobe therein having a slope greater than a predetermined minimum. A first pseudo-Gaussian sigma, calculated from the width and maximum amplitude of any such lobe, is compared with a second pseudo-Gaussian sigma, calculated from known conditions. If the first sigma is less than the second sigma, the lobe is deemed to be clutter related and is edited from the spectral envelope.