A procedure to predict the spectral resolution requirements of a Fourier Transform Infrared spectrometer

The DSTO employs a Bomem MR series Fourier Transform Infrared (FTIR) spectrometer for a variety of thermal signature measurements, passive remote sensing and spectroscopic applications. These measurements are often performed at the maximum resolution of the instrument with associated overheads in extended acquisition times and increased data size and storage issues. With many of the passive remote sensing and target detection applications envisaged for this and similar systems not requiring such high spectral resolution, a procedure to guide the relaxation and optimization of the spectral resolution of the system is required. Presented is a modeling procedure developed at DSTO that in part meets this objective. The approach employs empirical measurements of an absorbing plastic with a strong infrared absorption and then assesses the effects of broadening by the application of simple integration procedures. A simple Target to Clutter Ratio (TCR) and single band ´likelihood ratio detection´ analysis is presented. This illustrates the correlation between detection performance and spectral resolution. From this, the minimum spectral resolution for detection of the test example can be determined for the Bomem MR series spectrometer