Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Question on convolution

Status
Not open for further replies.

mngeow

New Member
Here is another question my teacher set for me but I can't solve:

**broken link removed**

Of course I could just z transform h(n) and x(n) but my teacher wants me to solve the question without doing so.

So far what I've got is,I changed the range of the summation from k=0 to infinity,I also know that for the h(k) part its u(k) since h(n)=u(n),and we can strike u(k) off as it will always be 1 since k does not go below 0.But after that point I'm kinda stuck,not too sure about the x(n-k) part.

Thanks alot! :D
 
Hi,

If you know what x(n) is you should know what x(n-k) is right?

For a different example, if you know that f(x) is x+1 and you want to know what f(m+x) is, you would end up with (m+x)+1.
 
I'm still abit unsure.But I'm guessing if thats the case then x(n-k) is 2(0.5)^(n-k) * u(n-k)? But if thats the case how do I continue from there?
I'm guessing you sub in values of K and it should be somehow equal to the geometric progression?
 
Last edited:
Hi again,

It looks like they want you to relate the 2*(0.5)^n part to the solution to the progression which is on the right.
It also looks like the sum parts will be limited to certain values depending on n, because u(n-k) is not always 1, which if it were, would make the series divergent, so u(n-k) sometimes going to 0 must make the convergence possible. So in other words it looks like the sum will only have to go between certain values of k rather than to infinity.

It also might make more sense if you do a couple little examples making n a particular value, like 1, 2, 3, 5 and 10 or something like that. You should see a pattern emerge.

You might also do a search for "Discrete Convolution".
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top