DFT z-transform help needed

Hi, I have the following questions that I am stuck on. Could you please
help in solving them or at least point me in the direction that can
help me? Thanks.

A given sequence x has x[n]=0 for n outside the range 0<=n<=10.
(a)The only computation that you can perform is one DFT. (Assume you
have a computer algorithm coded up to do a DFT, and you can use it
once). We want to find the value of X(e^j2pi7/12), where X(z) is the
z-transform of x[n]. What is the minmum length DFT that you would need
to perform, on what set of numbers, and how would you find the desired
result from it?(Show how you arrive at the answer)

(b) We now want the values of X(z) at the points z=e^j2pip/6 for
p=0,1,...,5. You are allowed to perform one DFT computation, and also
simple additions if needed. What is the minimum length DFT that you can
use, on what set of numbers, and how would you find the desired result
from it? (Justify your answer)

Thanks in advance.

farhatch@yahoo.com wrote:

> Hi, I have the following questions that I am stuck on. Could you please
> help in solving them or at least point me in the direction that can
> help me? Thanks.
> 
> A given sequence x has x[n]=0 for n outside the range 0<=n<=10.
> (a)The only computation that you can perform is one DFT. (Assume you
> have a computer algorithm coded up to do a DFT, and you can use it
> once). We want to find the value of X(e^j2pi7/12), where X(z) is the
> z-transform of x[n]. What is the minmum length DFT that you would need
> to perform, on what set of numbers, and how would you find the desired
> result from it?(Show how you arrive at the answer)
> 
> (b) We now want the values of X(z) at the points z=e^j2pip/6 for
> p=0,1,...,5. You are allowed to perform one DFT computation, and also
> simple additions if needed. What is the minimum length DFT that you can
> use, on what set of numbers, and how would you find the desired result
> from it? (Justify your answer)
> 
> Thanks in advance.
> 

(a) Ask your professor, the teaching assistant for the class, an 
assigned tutor or a friend.

I arrived at this answer by observing that the wording of the questions 
looks like they were copied out of a book or off of an exam sheet, and 
the assumption, based on a lack of grounding in any real-world problem, 
that you're not teaching the subject to yourself.

(b) Ask your professor, the teaching assistant for the class, an 
assigned tutor or a friend.

I justify this answer based on the fact that figuring out how to help 
someone through a problem like this without just telling the answer is 
work, and I'm stealing time away from my family to play on USENET right 
now.  Just telling you the answer would take significantly less time, of 
course, but it would be a profound disservice to you and your classmates.

If you _aren't_ taking a class then say so.  If you are at such a poor 
school or have such a poor professor that you can't get help then say so 
-- we'll all be much more sympathetic and helpful.

-- 

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

farhatch@yahoo.com wrote:
> Hi, I have the following questions that I am stuck on. Could you
please
> help in solving them or at least point me in the direction that can
> help me? Thanks.

These look way too similar to homework problems for anybody here to
make any sort of serious attempt to answer them for you.

> A given sequence x has x[n]=0 for n outside the range 0<=n<=10.
> (a)The only computation that you can perform is one DFT. (Assume you
> have a computer algorithm coded up to do a DFT, and you can use it
> once). We want to find the value of X(e^j2pi7/12), where X(z) is the
> z-transform of x[n]. What is the minmum length DFT that you would
need
> to perform, on what set of numbers, and how would you find the
desired
> result from it?(Show how you arrive at the answer)

I think this is a nicely posed question. Did you find it in a
textbook? If so, which one?

> (b) We now want the values of X(z) at the points z=e^j2pip/6 for
> p=0,1,...,5. You are allowed to perform one DFT computation, and also
> simple additions if needed. What is the minimum length DFT that you
can
> use, on what set of numbers, and how would you find the desired
result
> from it? (Justify your answer)

OK, it's almost christmas and I'll give you one hint: Zero padding.
Where does the length of the DFT enter the formulas? Can you tweak
the DFT length to fit the numbers that are given in the homework
assignment?

And no, it's no use flaming me for being arrogant or whatever.
This is so clearly a homework assignment that you will never get
any better answers here than I've already given you.

Rune

Yes these are h/w questions. Unfortunately, I signed up for this class
which was only one week long. I have no prior experience with DSP. The
other class that leads up to this course was cancelled by the school
and didn't bother to tell me that it would be very difficult to
understand the concepts if I didnt take the first course. To be honest
I had already given up and sent the professor an email that I will not
continue with it anymore; I was just curious if someone could answer it
since I spent the whole damn weekend trying to figure it out. I have
spent over $300 dollars in books over the last week but I think one
week for a 3 semester credit course is just not enough to catch up to
all the theory.

Anyway, I appreciate you guys taking time to answer, even though it
wasn't the answer to what I asked.

Cheers
Farhat

FC wrote:
> Yes these are h/w questions. [SNIP ;]

Don't give up yet ;)

I have observed that this group is not adverse to to assisting with
*HOMEWORK* as long as it is INITIALLY identified as such ;!

To obtain a "positive" response, demonstrate that you have tried to 
solve the problem on your own effort.

On 19 Dec 2004 12:55:13 -0800, "FC" <farhatch@yahoo.com> wrote:

>Yes these are h/w questions. Unfortunately, I signed up for this class
>which was only one week long. I have no prior experience with DSP. The
>other class that leads up to this course was cancelled by the school
>and didn't bother to tell me that it would be very difficult to
>understand the concepts if I didnt take the first course. To be honest
>I had already given up and sent the professor an email that I will not
>continue with it anymore; I was just curious if someone could answer it
>since I spent the whole damn weekend trying to figure it out. I have
>spent over $300 dollars in books over the last week but I think one
>week for a 3 semester credit course is just not enough to catch up to
>all the theory.
>
>Anyway, I appreciate you guys taking time to answer, even though it
>wasn't the answer to what I asked.
>
>Cheers
>Farhat
>

Hello Farhat,

   your sad story is hard to believe.
Actually, it's not believable (at least to me).

It hard to understand why you want the answers 
to your questions if you dropped out of 
the class.

You can convince us that your story is true if
you give us the name of your school and 
the Course Number of your DSP class, and your 
professor's name.

[-Rick-]

Rick,

Why don't you just say that you don't know how to solve this problem,
instead of trying to be a private detective? Maybe you are the stud on
this group and now feel ashamed that you can't figure out this problem.
There is no point in continuing this discussion as I received the
answers from my professor now. If I were you I would be very careful in
how I deal with people; the industry is pretty small and who knows you
may have to work for me someday. Understood, stud muffin?

Farhat

FC wrote:

> Rick,
> 
> Why don't you just say that you don't know how to solve this problem,
> instead of trying to be a private detective? Maybe you are the stud on
> this group and now feel ashamed that you can't figure out this problem.
> There is no point in continuing this discussion as I received the
> answers from my professor now. If I were you I would be very careful in
> how I deal with people; the industry is pretty small and who knows you
> may have to work for me someday. Understood, stud muffin?
> 
> Farhat

I actually have a bit of sympathy for the way you feel, but you couldn't
be further from the mark. You will find all the information you need in
http://www.amazon.com/exec/obidos/tg/detail/-/0201634678/104-1411742-8728737?v=glance
by you-know-who.

Jerry
-- 
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������

"FC" <farhatch@yahoo.com> wrote in message 
news:1103583199.531906.95710@f14g2000cwb.googlegroups.com...
> Rick,
> [...] If I were you I would be very careful in
> how I deal with people; the industry is pretty small and who knows you
> may have to work for me someday. Understood, stud muffin?

That's too funny.  And after Jerry's post, you'll be saying "Well, if I'd 
known that, then I wouldn't have been such an a**!".

Maybe you're thinking you should do more research next time, to ensure that 
such condescension is appropriate to the circumstances of the conversation.

Eventually, I'm sure you would figure out the real lesson on your own, but 
let me save you a few years.  This is it:  It's never appropriate.  Just 
don't be an a**.

--
Matt

Rick,
If you had such sympathy for me then why did you have to spit in my
face when I asked for help in this message board. I've heard of your
book and I've also heard that its a great book. I am sorry for calling
you whatever I called you, but it was just a natural reaction to you
and your friend's sarcasm and ganging up on me. I recognize that you
are a guru at your stuff but there was no excuse for you to behave like
and a** either. I will learn all about DSP with or without your help.
You can choose to not help someone, but its not your choice to put
someone down when they are asking for help. I still stand by my
position no matter what any of you or your buddies say. Just a friendly
advice for you; in the future, just ignore someone's request rather
than giving them a lecture or calling them a liar. It just saves
everyone a lot of grief, ok?

Farhat