The nonuniform discrete Fourier transform (NDFT) can be computed with a fast algorithm, referred to as the nonuniform fast Fourier transform (NFFT). In L dimensions, the NFFT requires O(N(-lne)L+(∏𝓁=1LM𝓁)∑𝓁=1LlogM𝓁) operations, where M𝓁 is the number of Fourier components along dimension 𝓁, N is the number of irregularly spaced samples, and e is the required accuracy. This is a dramatic improvement over the O(N∏𝓁=1LM𝓁) operations required for the direct evaluation (NDFT). The performance of the NFFT depends on the lowpass filter used in the algorithm. A truncated Gauss pulse, proposed in the literature, is optimized. A newly proposed filter, a Gauss pulse tapered with a Hanning window, performs better than the truncated Gauss pulse and the B-spline, also proposed in the literature. For small filter length, a numerically optimized filter shows the best results. Numerical experiments for 1-D and 2-D implementations confirm the theoretically predicted accuracy and efficiency properties of the algorithm.
[1]
Milton Abramowitz,et al.
Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables
,
1964
.
[2]
Vladimir Rokhlin,et al.
Fast Fourier Transforms for Nonequispaced Data
,
1993,
SIAM J. Sci. Comput..
[3]
G. Beylkin.
On the Fast Fourier Transform of Functions with Singularities
,
1995
.
[4]
Jaan Pelt.
Fast Computation of Trigonometric Sums with Applications to the Frequency Analysis of Astronomical Data
,
1997
.
[5]
C. O. H. Hindriks,et al.
Reconstruction of band‐limited signals, irregularly sampled along one spatial direction
,
1999
.