Questions tagged [convolution]

Questions on the (continuous or discrete) convolution of two functions. It can also be used for questions about convolution of distributions (in the Schwartz's sense) or measures.

Convolution is a commutative, associative, distributive operation between two functions that produces a third function. It is defined in the continuous domain as

$$(x \ast y)(t)=\int_{-\infty}^\infty{x(\tau)\space{}y(t-\tau)}\space{}d\tau$$

And in the discrete domain as

$$(x \ast y)[n]=\sum_{k=-\infty}^\infty{x[k]\space{}y[n-k]}$$

Its identity is the Dirac delta function $\delta(t)$ in continuous domain, and the Kronecker delta function $\delta[n]$ in discrete domain.

Reference: Convolution.

2979 questions
1
vote
1 answer

clarification asked for 'difference between convolution and crosscorrelation?'

I don't understand answer formulated in ways like this "Thus, $p\ast q$ is the distribution of $X+Y$. The cross-correlation $p\circ q$ is the distribution $c=(c_n)_n$ defined by $c_n=\sum\limits_kp_kq_{n+k}=P[Y-X=n]$ for every $n$. Thus, $p\circ q$…
user68610
  • 343
  • 2
  • 7
1
vote
0 answers

How can I interpret this convolution formula mentioned in a paper about systolic arrays?

I'm reading a paper about systolic arrays, the author mentioned this formula for the convolution and I cannot map it to the formula that I have in mind. What I can interpret here is as follows: I have to pad the smaller sequence with zeroes to…
1
vote
0 answers

Scaled version of convolution

I found that $$ \int f(t + x) g(x) dx = f \ast g^{-} (t), $$ where $g^{-}(x) = g(-x)$. My question is that following integral $$ \int f(t + \lambda x) g(x) dx, ~~~ \lambda \in (0, 1) $$ can be expressed in terms of convolution. Thanks in advance.
bakgu
  • 55
1
vote
2 answers

Output from convolving two functions with Dirac delta functions

I have two functions in a convolution problem, and I'm struggling to wrap my head conceptually around what I'm supposed to do. The functions are: \begin{align*} h(t) &= \delta(t-1) + \delta(t-3)\\ x(t) &= \delta(t-3) - 2\delta(t-4), \end{align*} and…
1
vote
2 answers

Graphical understanding of the convolution of discrete distributions.

I am studying convolution and trying to get a visual sense of the process. From the wikipedia page I understand what 2 continuous gaussian or 2 uniform distributions will produced when convolved, but I'm having a hard time determine what discrete…
1
vote
1 answer

Approximate Identity

Let ${({\varphi}_{n}})_{n=1}^{\infty}$ an Approximate Identity in Schwartz Space. Let $\alpha \in \mathbb{Z}^+$. Is it true or not the following statement? \begin{equation} \lim _{ n\longrightarrow \infty }{ \int _{ \mathbb{R} }^{ }{ { \left|…
Mani
  • 83
  • 7
1
vote
2 answers

How can you do convolution graphically?

I don't exactly remember whether I should get the common area under the curves of the functions being convolved or I should multiply them and get the area under the resulting curve.
1
vote
1 answer

What would the convolution of a time scaled function with a dirac delta function be?

I know the convolution: $x(t) * \delta(t-t_{0}) = x(t-t_{0})$. But what would the result be if I have a convolution: $x(\frac{t}{T}) * \delta(t-t_{0})$? (where $T \neq$ 0) Would it be $x(\frac{t}{T} - t_{0})$ or $x(\frac{t - t_{0}}{T})$?
Rish
  • 153
1
vote
0 answers

Calculating a two dimensional circular convolution using a two dimensional linear convolution

Assume we have a pair of two dimensional matrices $A$ and $B$. How can we calculate their circular convolution using only the linear convolution (and possibly padding numbers)?
rayes09
  • 51
  • 7
1
vote
1 answer

Why is convolution of these two function not commutative?

$m2(t)=z(t)*f(t)=\int^{\infty}_{-\infty}z(\tau)f(t-\tau)d\tau$ $\tau$ is a variable in place of $t$, which is now a constant. For $-1
1
vote
0 answers

Showing a particular convolution is associative

Supposing $f,g, h$ are defined on $\Bbb R$ and are all integrable on $\Bbb R$. Define their convolution as $f * g (z) = \int_{\Bbb R} f(x) g (z-x) dx$. Now I wish to show that this operation is associative, i.e. that $(f*(g*h))(x)=((f*g)*h)(x)$.…
1
vote
0 answers

Prove this property of convolution. d/dt y(t) = x(t) * [d/dt h(t)] and d/dt y(t) = [d/dx x(t)] * h(t)

I've tried solving for the first one: d/dt y(t) = x(t) * [d/dt h(t)] This is as far as I've gotten. integral from -inf to inf (x(tau) * d/dt h(t-tau) dtau) not sure how to go from here. d/dt y(t) = [d/dx x(t)] * h(t)
LaSzLo
  • 27
1
vote
0 answers

Maxium value of discrete convolution

I'm trying to calculate the maximum possible short-term energy $E[n]$ of a sampled signal $s$ in terms of $N$ and $\text{bitdepth}$. $$ E[n] =\sum_{m=-\infty}^{\infty} s^2[n]w[n-m] $$ where $$ w(n) = 0.54 - 0.46\; \cos \left ( \frac{2\pi n}{N-1}…
1
vote
1 answer

how to perform laplace transform of a convolution?

Can i say that the laplace transform of a function convoluted with itself $\left[\,\mathrm{f}\left(\,t\,\right)*\mathrm{f}\left(\,t\,\right)\,\right]$ is the square of the Laplace Transform of that…
1
vote
0 answers

Convolution theorem in finite domain

When ones tries to show the convolution theorem: $\int_{-\infty}^{\infty}\int_{-\infty}^{\infty}f(x') g(x-x')e^{-ikx}dx'dx = \int_{-\infty}^{\infty}dx'f(x') \int_{-\infty}^{\infty}g(x-x')e^{-ikx}dx $ then, it is made a change of variable, for…