Abstract

A systematic examination of the efficacy of window functions for reducing the spectral skirt of magnitude-mode Fourier transform spectra is reported. The efficacy is examined for the general case of a damped time-domain signal, with specific cases ranging from undamped to essentially completely damped signals. The choice of the optimal window is dependent upon the required dynamic range and the amount of damping in the time-domain data. For a dynamic range of less than 100:1 and moderate damping, the Hamming window is the window of choice. For larger dynamic ranges or greater damping, the 3-term Blackman-Harris window and the Kaiser-Bessel window are the windows of choice. The 3-term Blackman-Harris window is preferred for a dynamic range of 1,000:1 and the Kaiser-Bessel window is preferred for a dynamic range of 10,000:1. The sensitivity (signal-to-noise ratio) reduction for windows is reported for a damping range from zero to essentially complete damping. All windows examined have the same sensitivity reduction within 25%.

Apodizing functions for Fourier transform spectroscopy

David A. Naylor and Margaret K. Tahic
J. Opt. Soc. Am. A 24(11) 3644-3648 (2007)

Apodization and Interpolation in Fourier-Transform Spectroscopy*

A. S. Filler
J. Opt. Soc. Am. 54(6) 762-767 (1964)

Extension of the Norton–Beer apodizing functions in Fourier transform spectrometry

Lu Lee, He Chen, Siying Chen, and Yinchao Zhang
Appl. Opt. 51(20) 4622-4626 (2012)

Conventions and values for the multiplex advantage in Fourier transform spectroscopy

Tomas Hirschfeld
Appl. Opt. 16(12) 3070-3071 (1977)

Norton-Beer apodization and its Fourier transform

Konstantin F. F. Ntokas, Jörn Ungermann, and Martin Kaufmann
J. Opt. Soc. Am. A 40(11) 2026-2033 (2023)