H. Hauptman, A MINIMAL PRINCIPLE IN X-RAY CRYSTALLOGRAPHY - STARTING IN A SMALL WAY, Proceedings - Royal Society. Mathematical and physical sciences, 442(1914), 1993, pp. 3-12
The phase problem has been reformulated as the problem of finding the
constrained global minimum of a well-defined function of hundreds of v
ariables. The solution of this problem requires initial estimates of a
large basis set of phases. In this paper it is shown that random stru
ctures containing small numbers of atoms serve as surprisingly good st
arting models.