Tomographic phase diversity for phase retrieval on wide-field AO systems
Damien Gratadour(1), François Rigaut(2)
(1) LESIA, Observatoire Paris , (2) Gemini Observatory
Phase diversity is a commonly used technique to retrieve the wavefront at the focal plane. The usual algorithm involves two or more images of the same target with known phase changes like defocus. It has been shown to be very efficient at measuring on-axis the non-common path aberrations of classical AO systems. In this paper, we present an evolution of this algorithm towards tomographic measurements. This novel technique is dedicated to wide-field AO systems, allowing phase retrieval on multiple layers, conjugated at various altitudes. While the general grounds are very similar to classical phase diversity, the tomographic algorithm involves two or more images with known phase changes of several targets dispatched over the entire field of view. Regularization on the phase is usualy done by factorizing it on a basis of modes, traditionally Zernike polynomials. In this paper, we discuss the choice of a proper basis in the tomographic case and show that other basis such as disk harmonics are interesting alternatives in the case of real AO systems. We additionally propose two versions for this algorithm: an image-based and a Fourier-based both leading to comparable results. We finally present the results obtained on simulated data as well as on real data obtained on the Gemini MCAO system on which this algorithm has been used to estimate and compensate for non common path aberrations.