Microscanning in Hadamard spectroscopy.

Hadamard spectroscopy is a method of taking spectroscopic measurements suited to domains such as far infrared, where the sensor noise is dominant. Unfortunately, the signal-to-noise ratio of such a system scales poorly with the resolution. In this paper, we analyze a technique called microscanning that attempts to solve this problem by combining several lower resolution "preview" spectra to form a higher resolution spectrum. We provide a theoretical analysis of the method concerning the noise reduction and blurring. We conduct simulations based on the measured far infrared spectra of minerals in order to determine if the noise reduction can outweigh the loss in accuracy caused by the blurring effect, and we demonstrate the microscanning method in experimental conditions. We conclude that the microscanning method provides a noise reduction by a factor proportional to the number of previews and that it can be useful in far infrared spectroscopy.