Output of an LTI system when the input is multiplied with 'n'

Join the discussion
Registration is free. Start your own thread to ask a follow-up.
6 replies · 2K views
jaus tail
Messages
613
Reaction score
48

Homework Statement


upload_2018-2-6_17-14-21.png


Homework Equations


Find out the function of the system.
Apply that to input.

The Attempt at a Solution


From given input output, system is a differentiator.
So when input is n u[n]
output is differentiating it,
we get:
n (impulse response) + u[n]
Which is not in any of the option.
Book answer is B.
I'm struggling to get that.
 
Attachments
  • upload_2018-2-6_17-14-21.png
    upload_2018-2-6_17-14-21.png
    3 KB · Views: 1,050
Physics news on Phys.org
jaus tail said:

Homework Statement


View attachment 219792

Homework Equations


Find out the function of the system.
Apply that to input.

The Attempt at a Solution


From given input output, system is a differentiator.
So when input is n u[n]
output is differentiating it,
we get:
n (impulse response) + u[n]
Which is not in any of the option.
Book answer is B.
I'm struggling to get that.
The problem statement seems to be mixing its use of square brackets and curved brackets. Just so we're on the same page, I like to use square brackets when working with discrete time, so I'm going to use them here.

Also, I'm going to assume that [itex]\delta[n][/itex] is the Krockner delta function (which has a magnitude of 1 at n=0 and a magnitude of 0 elsewhere), which should not be confused with the Dirac delta function (which has infinite magnitude at n=0).

Code:
n      -3   -2   -1   0   1   2   3   4   5
u[n]    0    0    0   1   1   1   1   1   1
δ[n]    0    0    0   1   0   0   0   0   0

Now that we have that squared away, I think you are correct about the system being a differentiator.

So the first thing is to write out the sequence [itex]n u[n][/itex]. This is easy to calculate since you already know what [itex]n[/itex] is and what [itex]u[n][/itex] is.

Differentiate your [itex]n u[n][/itex] sequence, and what do you get?

[Edit: Also, for what it's worth, "differentiation" in continuous time becomes "difference" in discrete time. So can you find the difference equation for this system?]
 
Last edited:
  • Like
Likes   Reactions: berkeman
upload_2018-2-7_9-24-51.png
 
Attachments
  • upload_2018-2-7_9-21-24.png
    upload_2018-2-7_9-21-24.png
    1.9 KB · Views: 569
  • upload_2018-2-6_17-14-21-png.png
    upload_2018-2-6_17-14-21-png.png
    3 KB · Views: 556
  • upload_2018-2-7_9-24-36.png
    upload_2018-2-7_9-24-36.png
    36.7 KB · Views: 618
  • upload_2018-2-7_9-24-51.png
    upload_2018-2-7_9-24-51.png
    2 KB · Views: 419
Last edited:
jaus tail said:
How to find the difference equation?
From quora I found:
difference equation of f(n) = f(n+1) - f(n)
Close, but not quite perfect.

The equation does involve a difference between a couple of terms. Just make sure to get the right terms.
But here input is n u[n]
so how to find it's difference equation:
(n + 1) u [n + 1] - n u[n] ?

Start by writing out the sequence of [itex]n u[n][/itex]. Start at around n = -3 and work you way to around n = 5 or so.
 
  • Like
Likes   Reactions: jaus tail
Yeah I think i figured it out. I was editing my post when you replied. Is it right now?
 
  • Like
Likes   Reactions: collinsmark
jaus tail said:
Yeah I think i figured it out. I was editing my post when you replied. Is it right now?
Yes, that looks correct now to me. :smile:

(Edit: you skipped the number '4' in your n, but you seem to have worked it out correctly otherwise.)
 
  • Like
Likes   Reactions: jaus tail
Also gives me right answer of B. Thanks.
 
  • Like
Likes   Reactions: collinsmark