Editing Aperture masking interferometry (section)

==Description==
In the aperture masking technique, the [[speckle masking|bispectral analysis]] (speckle masking) method is typically applied to image data taken through masked apertures, where most of the aperture is blocked off and light can only pass through a series of small holes (subapertures). The aperture mask removes atmospheric noise from these measurements through the use of [[Closure phase|closure quantities]], allowing the bispectrum to be measured more quickly than for an un-masked aperture.

For simplicity the aperture masks are usually either placed in front of the [[secondary mirror]] (e.g. Tuthill et al.  2000) or placed in a re-imaged aperture plane (e.g. Haniff et al. 1987; Young et al. 2000; Baldwin et al. 1986), as shown in Figure 1.a) . The masks are usually categorised either as non-redundant or partially redundant. Non-redundant masks consist of arrays of small holes where no two pairs of holes have the same separation vector (the same '''baseline''' – see [[aperture synthesis]]). 

Each pair of holes provides a set of fringes at a unique spatial frequency in the image plane. Partially redundant masks are usually designed to provide a compromise between minimizing the redundancy of spacings and maximizing both the throughput and the range of spatial frequencies investigated (Haniff & Buscher 1992; Haniff et al. 1989). Figures 1.b)  and 1.c) show examples of aperture masks used in front of the secondary at the Keck telescope by Peter Tuthill and collaborators; Figure 1.b) is a non-redundant mask while Figure 1.c) is partially redundant. 

Although the signal-to-noise of [[speckle masking]] observations at high light level can be improved with aperture masks, the faintest limiting magnitude cannot be significantly improved for photon-noise limited detectors (see Buscher & Haniff 1993).