Wavepacket Detectors and Waveforms for Scale Separation of Low Altitude Targets from Clutter

Subclutter visibility may be achieved using wavelets and wavepacket methodss. Wavelets offer the ability to separate features based on scale. Wavepackets and the best-basis algorithm offer the further advantage of adaptability; bases adapt to the signal to provide optimum signal representation. We propose to develop techniques for separting low altitude, low observable targets from clutter on the basis of scale differences. This approach provides subclutter visibility even when the target doppler exists in the clutter region. Therefore, when Fourier methods cannot distinguish target from clutter by frequency, wavepacket methods should separate target from clutter by scale. We will investigate the benefit gained by using the translation invariant wavelet transform (TIWT), developed at Aware, Inc., in the best-basis wavepacket algorithm. the TIWT provides better signal representations than those resulting from the standard wavelet transform. We will characterize clutter probability density functions (pdfs) and parameters through the best-basis wavepacket representation. We intend to develop wavepacket detectors based on likelihood ratio approaches. We will develop a wavepacket based CFAR detector and compare performance to Fourier based approaches. We will develop wavepacket based waveforms to separate from clutter by scale.