Robust interferometric imaging via prior less phase recovery: redundant spacing calibration with generalized closure phases
Abstract
Redundant Spacing Calibration (RSC) techniques employ redundancy in the baselines of a telescope array to eliminate the contribution of atmospheric turbulence in the interferometric observables. Whereas conventional techniques for this elimination require the enforcement of prior constraints on the underlying image, RSC algorithms can be, in principle, mathematically well-posed and hence require no such prior knowledge. Traditionally these algorithms have been applied directly to the fringe measurements. However, in scenarios of low photon flux, such as those arising in the observation of dim objects in space, single-exposure fringe measurements are not reliable observables in general. Instead one must rely on time-averaged, atmosphere-invariant quantities such as the bispectrum. In this paper, we develop a novel algorithm for redundant arrays which provides robust image reconstruction using integrable atmosphere-invariant observables. Our algorithm utilizes standard linear estimation methods, as well as techniques from lattice theory, to reliably estimate the Fourier phase. Moreover, we provide theoretical and empirical evidence that generalizing the classical bispectrum to higher-order atmosphere-invariant observables, which we call n-spectra, can offer significant performance gains.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 14, 2016
- Accession Number
- AD1032985
Entities
People
- Binoy G. Kurien
- Jonathan B. Ashcom
- Vahid Tarokh
- Vinay N. Shah
- Yaron Rachlin
Organizations
- MIT Lincoln Laboratory