Time-domain weighted spectral matching by nonlinear wavelength sweeping

This paper proposes a method for enhancing the small difference between two reference spectra on the spectral matching imager. This method consists in nonlinear sweeping of the peak wavelength of the monochromatic light. Theoretical analysis reveals that the nonlinear wavelength sweeping results in weighted spectral correlation. An optimization procedure is described for obtaining the weight function that brings the spectral correlation between two similar spectra as close to zero as possible. Simulation results demonstrate that the proposed method can enhance the small difference between two reference spectra.