20 October 2004 General fringe decomposition and statistical bias correction in optical interferometry
Author Affiliations +
Interferometric fringes are traditionally decomposed using the Discrete Fourier Transform (DFT). However, the application of the DFT is only correct in cases where the fringes are sampled evenly with delay and over integer number of fringe periods. This ideal case is often not achieved in Optical Interferometry. Fringe spectrography, non-linear fringe sweeps and image-plane beam combiners are typical cases of where the DFT approach fails to make most efficient use of the data. The authors assert that in many cases alternative and more efficient fringe decompositions exist but which may exhibit considerably different noise behaviour to the DFT. The authors present the mathematical results important for correcting for statistical bias in the powerspectrum and bispectrum constructs of a completely general fringe decomposition. An estimator for the noise in the powerspectrum has also been derived. The authors believe this to be the first analytical derivation of statistical bias and noise in interferometry that treats both photon counting noise as well as Gaussian read out noise.
© (2004) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Hrobjartur Thorsteinsson, David Felix Buscher, "General fringe decomposition and statistical bias correction in optical interferometry", Proc. SPIE 5491, New Frontiers in Stellar Interferometry, (20 October 2004); doi: 10.1117/12.550983; https://doi.org/10.1117/12.550983

Back to Top