Solved: Heavyside Convolution: Calculate (\theta \ast \theta)(x)

  • Thread starter Thread starter Sigurdsson
  • Start date Start date
  • Tags Tags
    Convolution
Click For Summary

Homework Help Overview

The discussion revolves around calculating the convolution of the Heaviside step function, denoted as (\theta \ast \theta)(x). Participants are exploring the implications of the convolution definition and the behavior of the Heaviside function across different intervals.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • The original poster attempts to compute the convolution but expresses confusion over the outcome, particularly regarding the integration limits and the resulting expression. Other participants question the validity of the integration steps and suggest clarifying the limits of integration.

Discussion Status

Participants are actively engaging with the problem, offering insights into the correct limits of integration and the behavior of the Heaviside function. Some have provided clarifications that help guide the original poster towards a better understanding of the convolution process.

Contextual Notes

There is a focus on the conditions under which the Heaviside function is non-zero, and participants are examining the implications of these conditions on the convolution integral. The original poster expresses a preference for rigorous methods over intuitive reasoning, highlighting a potential tension between different approaches to solving the problem.

Sigurdsson
Messages
24
Reaction score
1

Homework Statement


Calculate the convolution
[tex](\theta \ast \theta)(x)[/tex]

Homework Equations


Convolution is defined as:
[tex](f \ast g)(x) \equiv \int_{-\infty}^{\infty} f(x - y) g(y) \ dy = \int_{-\infty}^{\infty} f(y) g(x-y) \ dy[/tex]

The Attempt at a Solution


I know this is probably easy for many but I'm really baffled with the outcome. Here we go

[tex](\theta \ast \theta)(x) = \int_{-\infty}^{\infty} \theta(x - y) \theta(y) \ dy = \int_{-\infty}^0 \theta(x - y) \underbrace{\theta(y)}_0 \ dy + \int_0^{\infty} \theta(x - y) \underbrace{\theta(y)}_1 \ dy[/tex]
[tex]= \int_0^{\infty} \theta(x - y) \ dy = y \theta(x - y)[/tex]

However my result should be
[tex]= y \theta(y)[/tex]

Gawdemmit, I can't spot my fault here. What am I missing?
Cheers.
 
Physics news on Phys.org
Your integration with respect to y can't end up with a y in the result. How are you getting to your final result?
 
You're right, It should be

[tex]\int_0^y \theta(x - y) dy[/tex]
 
The upper limit should be x, not y, because the step function equals 1 only when x-y > 0. Remember y is a dummy variable in the convolution integral. You should end up with a function of x.
 
[tex] \theta \ast \theta (x) = \int_{-\infty}^{\infty}{\theta(x - y) \, \theta(y) \, dy}[/tex]
Look where the argument of each of the functions becomes zero:
[tex] y = 0, \ y = x[/tex]
So, if [itex]x < 0[/itex], the intervals in question are:

[tex] \begin{array}{c|c|c|c|c|c|c|c|}<br /> & -\infty & & x & & 0 & & \infty \\<br /> \hline<br /> \theta(y) & & - & - & - & 0 & + & \\<br /> \hline<br /> \theta(x - y) & & + & 0 & - & - & - &<br /> \end{array}[/tex]

Since the Heaviside step function is zero whenever its argument is negative, we must have to "+" in both rows for the product to be non-zero. As you can see, this is never the case for [itex]x < 0[/itex].

Now, suppose [itex]x > 0[/itex]. The intervals in question are:
[tex] \begin{array}{c|c|c|c|c|c|c|c|}<br /> & -\infty & & 0 & & x & & \infty \\<br /> \hline<br /> \theta(y) & & - & 0 & + & + & + & \\<br /> \hline<br /> \theta(x - y) & & + & + & + & 0 & - &<br /> \end{array}[/tex]
Now, the only interval when the product is non-zero is [itex](0, x)[/itex]. Thus, the convolution integral reduces to:
[tex] \theta \ast \theta(x) = \left\lbrace \begin{array}{ll}<br /> 0 &, x < 0 \\<br /> <br /> \int_{0}^{x}{d y} &, x > 0<br /> \end{array}\right.[/tex]

I think it is pretty straightforward to evaluate the remaining steps, and use the definition of Heaviside's step function to encompass both cases under a single expression.
 
Thanks guys I got it now.

The way you laid it out Dickfore makes it really clear but it unnerves me to solve an integral like that using just simple logic. I usually go straight for some rigorous integration rules and tricks instead of just visualizing the problem.

Thanks again
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 19 ·
Replies
19
Views
3K
Replies
19
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 1 ·
Replies
1
Views
1K
Replies
4
Views
2K