Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Convolution intergral - dumb question

  1. Oct 23, 2008 #1
    OK. I understand almost everything I need to know in order to do a convolution problem except how to set up what needs to be integrated. The way my professor does it is:

    1) flip the impuse
    2) move h(t) over x(t) - we usually do this in 5 different regions
    3) for each region the bounds have to be chosen and this I can do but how do I figure out what needs to be integrated?


    for this problem you have
    0<t<(t-1) and the limits of integration are from 0 to t but what needs to be integrated.

    I don't know if I'm getting caught up on the math or what but I do know that I'm completly lost with this. If someone could explain it to me in detail I would greatly appreciate it. I have a test tomorrow and don't want to fail it.
  2. jcsd
  3. Oct 24, 2008 #2
    Hi, if the amplitude of the impulse is A, and the slope is line equation is y=mx+b (m is the slope, b is 0), you must integrate A*mx from 0 to t. In this case you would get (A*m*t^2)/2

    Try the convolution of two impulses, one with amplitude A and another with B.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook