Scale floating point numbers. IEEE scalbn().
x = pow2(y) computes 2 to the power y.
x = pow2(f,e) for a real matrix f and a matrix of integers e
computes x = f .* (2 .^ e). The result is computed quickly
by simply adding e to the floating point exponent of f.
This corresponds to the ANSI C function ldexp() and the IEEE
floating point standard function scalbn().
See also LOG221741AB, REALMAX , REALMIN .