While dispersive spectrometers are a fairly direct way to separate optical radiation into its constituent irradiance spectrum, there are a few applications for which alternate imaging spectrometer designs are advantageous. In particular, this chapter describes a class of instrument designs that employs interferometry to produce signal modulation on a per-pixel basis related to the irradiance spectrum by a Fourier transform. Such an instrument is called a Fourier transform spectrometer (FTS) and requires digital processing to compute a representation of the irradiance spectrum from measured data. These spectrometers are also called Fourier transform infrared (FTIR) spectrometers, because they have historically been used in MWIR and LWIR spectral regions to capture extremely high-resolution spectra of gaseous materials. They are able to do so because their spectral resolution is limited only by mechanical parameters and not fundamental component characteristics such as prism material dispersion and grating periods. Laboratory instruments have been developed that achieve stability over very large interferometer baselines, producing the excellent resolution needed to resolve the fine rotational features of these materials. Traditionally, such instruments have been nonimaging; however, they have recently extended to imaging operation to support specific hyperspectral remote sensing applications, where their high achievable spectral resolution and natural staring imaging format are advantageous (Wolfe, 1997). The theoretical treatment of FTS operation is initiated with a discussion of traditional, nonimaging systems but is then extended to particular issues concerning imaging operation, along with a description of some examples of systems.
Fourier Transform Spectrometer Design and Analysis
Digital signal processing