Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

The inner product of the two-dimensional sequences $f(x,y)$ and $g(x,y)$ is equal to the inner product of their Fourier transforms, that is:

$$\sum_{x=-\infty}^{\infty}\sum_{y=-\infty}^{\infty}f(x,y)g^*(x,y)=\dfrac{1}{4\pi^2}\int_{-\pi}^{\pi}\int_{-\pi}^{\pi}F(w_x,w_y)G^*(w_x,w_y)\,dw_x\,dw_y.$$

I am trying use a Fourier transform inverse and follow re-arranged the integrals and use the Dirac function. But I don't know Why the integrals have limits $(-\pi,\pi)$.

share|improve this question
    
Have you tried with Bessel-Parseval's inequality? –  Davide Giraudo Nov 12 '12 at 15:04
    
I know Parseval's theorem and Bessel inequality but Bessel-Parseval's where I will be able to found? –  juaninf Nov 12 '12 at 15:15
1  
I meant the result which gives you an isometry between $L^2((-\pi,\pi)^2)$ and $\ell^2(\Bbb Z^2)$ (I actually don't know exactly the name). Then we polarize. –  Davide Giraudo Nov 12 '12 at 15:17
add comment

1 Answer

The set $\{e^{imx}e^{iny},m,n\in\Bbb Z\}$ forms a Hilbert basis of the space $H_1:=L^2((-\pi,\pi)^2)$ with the canonical inner product. Denote $e_{m,n}$ the sequence whose all terms are $0$, except the $(m,n)$-th which is $1$. Then $\{e_{m,n},m,n\in\Bbb Z\}$ form a Hibert basis for $H_2:=\ell^2(\Bbb Z^2)$.

The two mentioned Hilbert spaces are isometric (say $\iota\colon H_1\to H_2$, with $\iota(e^{imx}e^{iny})=e_{m,n}$), so for each $x\in H_1$, $\lVert \iota(x)\rVert_{H_2}²=\lVert x\rVert_{H_1}^2$ . Such an equality is true for $x\pm iy$ and $x\pm y$, which gives $$\langle \iota(x),\iota(y)\rangle_{H_2}=\langle x,y\rangle_{H_1},$$ what is wanted.

share|improve this answer
    
thanks by your reply. I don't understand your comment "Such an equality is true for $x\pm iy$ and $x\pm y$" to obtain $\langle \iota(x),\iota(y)\rangle_{H_2}=\langle x,y\rangle_{H_1},$ from $\lVert \iota(x)\rVert_{H_2}^2=\lVert x\rVert_{H_1}^2$. Which are the intermediary steps? –  juaninf Nov 12 '12 at 18:03
    
You have a Hermitian form $q$ such that $\lVert q(u)\rVert=\lVert u\rVert$ for all $u$ (that is, it conserves the norm) and you want to show it also conserves the inner product. What does $\lVert x+iy\rVert^2+\lVert x-iy\rVert^2$ give? –  Davide Giraudo Nov 12 '12 at 20:27
add comment

Your Answer

 
discard

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.