I've been trying to find a proof of the following, but still I m unable to proof it, can someone help me? $$ ℱ[x(t)g(t)] = \frac{1}{2\pi} [X(\omega)*G(\omega)] $$ enter image description here

  • $\begingroup$ You can’t have looked very hard for the proof. Almost any text on signal processing will do. $\endgroup$ Jun 7, 2020 at 22:18
  • 1
    $\begingroup$ @DilipSarwate O dear, Can you suggest me text that has proof of this? $\endgroup$ Jun 8, 2020 at 6:41
  • $\begingroup$ Try replacing only $x(t)$ by its representation as the inverse Fourier transform of $X(\omega)$ so that the integral formula for $X_q(\omega)$ involves only two integrals instead of the three that you have. Then interchange the order of integration. When you finish solving the question, write up a nicely LaTexed answer and post it. Yes, it is permissible (even encouraged) to post an answer to one's own question. $\endgroup$ Jun 8, 2020 at 13:52

2 Answers 2


I am very thankful to Dilip-sarwate and Gilles, who took their precious time to understand my problem and guide me.
So, Now I'm going to write the correct solution to my question. Which is as follows: $$ ℱ[x(t)g(t)] = \int_{t=-\infty}^{\infty}[x(t)g(t)]e^{-j\omega t}dt $$
As we know : $$ x(t) = \frac 1{2\pi}\int_{\alpha=-\infty}^{\infty}X(\alpha)e^{j\alpha t}d\alpha $$
Therefore, $$ ℱ[x(t)g(t)] = \int_{t=-\infty}^{\infty}\bigg[\frac 1{2\pi}\int_{\alpha=-\infty}^{\infty}X(\alpha)e^{j\alpha t}d\alpha\bigg]g(t) e^{-j\omega t}dt $$
After rearranging, we have $$ ℱ[x(t)g(t)] = \frac 1{2\pi}\int_{\alpha=-\infty}^{\infty}X(\alpha)\bigg[\int_{t=-\infty}^{\infty}g(t)e^{-j(\omega-\alpha)t}dt\bigg]d\alpha $$
As we also know that : $$ G(\omega) = \int_{t=-\infty}^{\infty}g(t)e^{-j\omega t}dt $$ and replacing $\omega$ with $\omega-\alpha$, we get: $$ G(\omega-\alpha) = \int_{t=-\infty}^{\infty}g(t)e^{-j(\omega-\alpha)t}dt $$
Therefore, $$ℱ[x(t)g(t)] = \frac 1{2\pi}\int_{\alpha=-\infty}^{\infty}X(\alpha)G(\omega-\alpha)d\alpha$$
Finally, we get $$ ℱ[x(t)g(t)] = \frac 1{2\pi}[X(\omega)*G(\omega)] $$
[$\because$ $x_1(t)*x_2(t) = \int_{\tau=-\infty}^{\infty}x_1(\tau)x_2(t-\tau)d\tau$]

  • $\begingroup$ Good job. After a couple of days, you can even accept your own answer as the most satisfactory answer. $\endgroup$ Jun 9, 2020 at 0:49
  • $\begingroup$ You're welcome, well done! $\endgroup$
    – Gilles
    Jun 9, 2020 at 7:11


Write it as: $$ \mathcal{F}\big\{x_1(t)x_2(t)\big\} = \frac 1{2\pi}\int_{-\infty}^{\infty}\bigg[\mathcal{F}^{-1}\big\{X_1(\alpha)\big\}x_2(t)\bigg]e^{-j\Omega t}dt $$

You correctly used the inverse Fourier transform of $X_1(\alpha)$ for a new expression of $x_1(t)$ as in $(1)$

$$ x_1(t) = \frac 1{2\pi}\int_{-\infty}^{\infty}X_1(\alpha)e^{j\alpha t}d\alpha\tag{1} $$ Don't express $x_2(t)$ in the form of the inverse Fourier transform, leave it as it is and you'll end up with an exponential to the power $-j(\Omega - \alpha)t$. The expression can be further simplified and give equation $(2)$ $$ \frac 1{2\pi}\int_{-\infty}^{\infty}X_1(\alpha)X_2(\Omega - \alpha)d\alpha\tag{2} $$


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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