Detection performance of the fractional Fourier transform (chirp FFT) for frequency modulated signals

Locke, Jonathan D. and White, Paul R. (2009) Detection performance of the fractional Fourier transform (chirp FFT) for frequency modulated signals. Papdakis, J.S. and Bjorno, L. (eds.) In Proceedings of the Third International Conference on Underwater Acoustic Measurements, Technologies and Results. Foundation for Research and Technology - Hellas. 6 pp .

Abstract

The Fractional Fourier transform (FrFT) is well suited to the processing of linear frequency modulated signals. Amongst the abilities it potentially offers are: the detection of signals at lower SNRs, better signal separation and the possibility of estimating the chirp rate of a signal. This work concentrates on using such a transform for detecting frequency modulated signals. For the detection problem the magnitude of the FrFT is sufficient and consequently the algorithm can be implemented efficiently by modifying a standard FFT, this implementation we shall refer to as a ‘Chirp-FFT’. This study considers the performance of the Chirp-FFT routine compared to the FFT routine for the detection of quiet underwater signals of both biological and man-made origin whose chirp rates are unknown a priori. Critically, the performance analysis here includes the effect the algorithms have on the background noise – a factor overlooked in other similar studies. The comparison is made on the basis of real data. It is shown that, unsurprisingly, the detection advantage offered by the Chirp-FFT is very much dependent on the character of the signal, but for the examples considered the gain in performance is in the range 0 - 3 dB

This record has no associated files available for download.

Contributors

Author: Paul R. White

Download statistics