BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:An algorithm for the denoising of cosmic strings in CMB data - Yve
 s Wiaux (EPFL\, Lausanne\, Switzerland)
DTSTART:20080708T153000Z
DTEND:20080708T163000Z
UID:TALK12798@talks.cam.ac.uk
CONTACT:Jane Buckle
DESCRIPTION:Topological defects arise from early phase transitions in cosm
 ological\nscenarios predicted by theories for the unification of the funda
 mental\ninteractions. The current status of observation leaves room for th
 e\nexistence of such defects. The issue of their detection thus\nrepresent
 s today a central question in cosmology.\n\nIn this talk we present the fi
 rst version of an efficient algorithm\nfor the denoising of a non-Gaussian
  string signal in the standard\nGaussian CMB signal seen as additive noise
 . A steerable wavelet\ndecomposition allows one to probe the signal at var
 ious analysis\nscales and orientations in terms of wavelet coefficients. T
 he wavelet\ncoefficients of the string signal are estimated by a Bayesian 
 least\nsquare denoising procedure relying on a string model trained on\nsi
 mulations\, and on a preliminary assessment of the string\ntension. The st
 ring signal itself is reconstructed from its identified\ncoefficients and 
 the string network can then be mapped as the\nmagnitude of gradient of the
  denoised signal. After denoising\, the\nkurtosis of the magnitude of grad
 ient can be used as a statistic to\ndefine a hypothesis test for the exist
 ence of strings.\n\nWe discuss the performance of the algorithm in the per
 spective of\nforthcoming arcminute CMB experiments. Further improvements a
 re also\ndiscussed\, both to provide fine constraints on the string tensio
 n\nthrough denoising\, and to deal efficiently with secondary CMB\nanisotr
 opies.\n
LOCATION:Ryle Seminar Room\, Cavendish Laboratory
END:VEVENT
END:VCALENDAR