okay, so following Seraj and implementing problem 1(b), I get the following for a Gaussian RV in C:

(ln(1-drand48())/(-0.5))^.5*cos(drand48())

Where the whole part in front of the cosine is the square root of an exponential RV (lambda = 0.5); which was obtained from a continuous RV over [0,1] (represented as the drand48() parameter) and using 1(b) as a guideline for obtaining such a RV.

Alumni Liaison

Ph.D. on Applied Mathematics in Aug 2007. Involved on applications of image super-resolution to electron microscopy

Francisco Blanco-Silva