# constrained optimization of dot product

Given a real matrix $A$ find a positive vector $x$ of unit length ($x^T x = 1$) for which $x^T A^T A x$ is minimal (closest to $0$).

A has size about $1000 \times 20$ and can be written as $[ A_P | A_N ]$ ($A_P$ contains only positive and $A_N$ only negative numbers)

Is there any deterministic algorithm which can reach the solution (global minimum) in polynomial time?

-
Isn't this a least squares problem? Weighted_least_squares –  nbubis Apr 10 '12 at 11:59
I think the problem is in the constraint. Without the constraint $x^Tx = 1$, there is trivial solution $x=0$ –  Boris Apr 10 '12 at 14:34