Abstract
The central limit theorem of statistics states that the arithmetic average of N (finite) random variables obeys a normal probability law. A requirement of the theorem is that the probability density common to the random variables have finite variance. This is not the case in optics, where for example the probability density on a photon position xi in an image can obey a sinc2 x law. The variance of this law is infinite. Then what probability density is obeyed by the arithmetic average of the xi in the optical case? It is shown that the law is Cauchy, i.e., of the form , a = const. Surprisingly, both the Cauchy form of the law and its free parameter a are entirely independent of the aberrations of the imaging system.
© 2003 Optical Society of America
PDF ArticleMore Like This
Juan Correa, Enrique Manzano, Ryan Tracy, and John R. Thompson
IThG37 International Quantum Electronics Conference (IQEC) 2004
Roland Winston
MDD1 Frontiers in Optics (FiO) 2003
C. Hesse, C. Dan, D. Wolfersberger, N. Fressengeas, and G. Kugel
413 Photorefractive Effects, Materials, and Devices (PR) 2003