Time-frequency representation of wideband radar echo using adaptive normalized Gaussian functions

We apply a signal representation scheme that uses adaptive normalized Gaussian functions, introduced by Qian and Chen (see Signal Processing, vol.36, no.1, p.11, 1994), to the analysis of wideband radar echo comprised of scattering centers and interior resonances. This scheme provides an adaptive adjustment of the time and frequency resolution to best match the signal, in contrast to the Gabor expansion and wavelet decomposition which use preselected, fixed resolution cells. This representation is economical and allows an easy and meaningful energy representation in the time-frequency domain.