On a pseudo-subspace framework for discrete Fractional Fourier transform based chirp parameter estimation

Sinusoids remain the prototypical waveform for signal modeling, analysis, detection, and estimation in stationary environments, but are unsuitable for the analysis of signals with non stationary frequency content. In, the MA-CDFRFT was introduced as a useful tool for the analysis of multicomponent chirp signals in the absence of noise. Subspace approaches derived from a eigenvalue decomposition of the correlation matrix of noisy observations of sinusoidal signals, such as the MUSIC or minimum-norm algorithms are popular approaches for estimating the parameters of multiple sinusoidal signals in white noise. In this paper, we extend the MA-CDFRFT methodology to develop a pseudo-subspace approach towards chirp parameter estimation.