Complex Generalized Gaussian distribution generator

This code generates circular-complex samples that follow a generalized gaussian distribution.

Generates complex generalized gaussian random variables with augmented
covariance matrix Ta = [2*s 0; 0 2*s];
and shape parameter c, where c = 1 corresponds to the Gaussian case.
x = cggd_rand(c,s,N) generates a vector 1xN of complex samples with a circular
gaussian distribution with shape parameter c, and variance 2*s.
[x,xa] = cggd_rand(c,s,N) generates an additional augmented matrix 2xN,
xa = [x;conj(x)].
The samples are generated according to the results presented in:
Mike Novey, T. Adali, and A. Roy "A complex generalized Gaussian Distribution---
characterization, generation, and estimation" in
IEEE Trans. Signal Proc., Vol 58. No 3, MARCH 2010