# How to prove a function is the Fourier transform of another $L^{1}$ function?

If $m(\xi)$ satisfies $$D^{\alpha}m(\xi)\leq \frac{C}{(1+|\xi|)^{|\alpha|+1}}$$ then is $m$ a Fourier transform of a $L^{1}$ function? (Note that the Bernstein theorem can't be applied here, since $m(\xi)$ may not be in $H^{s}$, where $s>\frac{n}{2}$.)

Generally, are there some simple ways to make sure that a given function belongs to $\mathcal{F}L^{1}$?

-
Are you assuming the inequality for all $\alpha$? Wouldn't that imply that $m$ is smooth and all its derivatives are integrable? –  timur Sep 4 '12 at 22:40
@timur:I think it should be $\alpha\leq [\frac{n}{2}]$ –  sun Sep 5 '12 at 13:12

There is a well known class of functions known as Schwartz class. In this class, the functions have the property that they and their derivatives tends to zero as $|x|\rightarrow \infty$, faster than any positive power of $x^{-1}$, or in other words, suppose that for each positive integer $N$ and $n$, $$\lim_{|x|\rightarrow \infty} x^N g^{(n)}(x) = 0\,.$$
Also, these kind of functions are known as good functions. For example, $x^m {\rm e}^{-x^2}$ is a good function. One of the properties of these functions is
$$|f(x)| < C \frac{1}{(1+|x|)^m}$$ for any $m \in N \,.$
Right,$\mathcal{FS}$ belonngs to any $L^{p}$,$1\leq p \leq \infty$,and it's well known that $\mathcal{F}L^{1}\subset C_{0}$,so I'm very interested of how to identify a fuction(together with it's derivatives) of some decay at $\infty$ is the fourier transform of a integral function(see the example above) –  sun Aug 28 '12 at 9:00