A sum of squares approximation of nonnegative polynomials

We show that every real nonnegative polynomial $f$ can be approximated as closely as desired by a sequence of polynomials $\{f_\epsilon\}$ that are sums of squares. Each $f_\epsilon$ has a simple et explicit form in terms of $f$ and $\epsilon$. A special representation is also obtained for convex polynomials, nonnegative on a convex semi-algebraic set.