MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Considering the function $f:\mathbb{R} \to \mathbb{C}$, with $\left| f(x) \right|=1$ for all $x\in \mathbb{R}$.
Considering $g:\mathbb{R} \to \mathbb{C}$ with $\int_{-\infty}^{\infty}{\left|g(x)\right|^2dx}=1$
I am interested on properties of the amplitude of the Fourier Transform of the product of $f$ and $g$:
Is there any constraint on $A$ apart from the fact that A is real positive? Considering a fixed $g$, is it possible to attain any $A$ simply by changing $f$?
Thank you for any help

share|cite|improve this question
up vote 5 down vote accepted

If $g$ happens to be in $L^1$, then the amplitude of the Fourier transform of $fg$ is bounded by the $L^1$ norm of $g$, for any unimodular $f$. This is the only restriction from above since you can always choose $f$ so that $fg\ge 0$, thus bringing the (essential) supremum of $\widehat{fg}$ up to $\|g\|_{L^1}$.

Another part of the question is how small we can make $A$. I guess "arbitrarily small", but don't have a proof. (Except for special case: if $g$ is in $L^1$, then we can chop it into pieces with disjoint supports and small $L^1$ norm, and then use $f$ to move the Fourier transforms of pieces far from one another.)

share|cite|improve this answer
I have a special case of this lower bound problem: More precisely, let f(x)=\exp(i \theta(x)) and assume |\theta'(t)|<= \Omega and f(0)=f(a)=1. Define [the windowed FT] A(k)=\int_0^a \exp(i k t) f(x) dx How does a lower bound for [L2 norm on a fixed interval] \int_0^k_c |A(k)|^2 dk scale with \Omega? I suspect that the lower bound should be O(1/\Omega^2) or something similar but I am often wrong. Maybe I should pose this as a separate question? – Kaveh Khodjasteh Feb 23 '10 at 23:05

The only thing you can say is $\int_{-\infty}^{+\infty}|A(k)|^2dk=1$, since the Fourier transform is an isometry of $L^2(\mathbb{R})$.

Any function $A(k)$ with $\int_{-\infty}^{+\infty}|A(k)|^2dk=1$ is the Fourier transform of a function of the form $f\cdot g$. Let $h$ be the inverse Fourier transform of $A(k)$, so that $A=\hat h$ and $\int_{-\infty}^{+\infty}|h(x)|^2dx=1$. We can write $h=g\cdot f$ with $g=h/f$; then $$ \int_{-\infty}^{+\infty}|g(x)|^2dx=\int_{-\infty}^{+\infty}|h(x)|^2dx=1 $$ and $A=\widehat{f\cdot g}$

share|cite|improve this answer
I misunderstood the question. The size of the Fourier transform of g at infinity is "determined" by the smoothness of g. Multiplying by f does not change the smoothness, so I would not expect a change in the size of |A(k)|. – Julián Aguirre Feb 10 '10 at 16:14
If we add the hypothesis that everything is $C^\infty$-smooth, including f, does it help? And how would you go to find the right f for given functions A and g? Thank you a lot for your help – Jean-François Morizur Feb 11 '10 at 10:01

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.