Title :
Signal-adapted wavelets for Doppler radar system
Author_Institution :
Sch. of Eng. & Tech., Deakin Univ., Geelong, Vic., Australia
Abstract :
In this paper, a method based on genetic algorithms is proposed for constructing signal-adapted wavelets for the wavelet-based ammunition Doppler radar system. In the wavelet-based Doppler system, the joint time-frequency analysis of the Doppler signals is done using the wavelet packet best-basis transform (WPBBT). Either standard wavelets from the literature or specially constructed signal-adapted wavelets can be used in the WPBBT. With signal-adapted wavelets the choice of wavelets for the WPBBT will not be limited to the standard wavelets from the literature. The best bases entropy is a performance measure of the wavelet when it is used in the WPBBT and is used as the cost functional to be minimised in the genetic algorithm. The proposed method can be used to construct signal-adapted wavelets for other wavelet-based applications such as image compression.
Keywords :
Doppler radar; genetic algorithms; time-frequency analysis; wavelet transforms; Doppler radar system; Doppler signals; best-basis entropy; genetic algorithms; image compression; joint time-frequency analysis; signal-adapted wavelets; wavelet packet best-basis transform; Costs; Doppler radar; Encoding; Entropy; Filters; Genetic algorithms; Libraries; Nonlinear equations; Time frequency analysis; Wavelet packets;
Conference_Titel :
Control, Automation, Robotics and Vision, 2002. ICARCV 2002. 7th International Conference on
Print_ISBN :
981-04-8364-3
DOI :
10.1109/ICARCV.2002.1234783
