Complex Summation

1. Jul 13, 2009

WolfOfTheSteps

This is not really a homework problem, but I'm studying a text, and I came across this:

http://img198.imageshack.us/img198/4586/sumh.jpg [Broken]

I know how to get that fraction with the exponents in it (using a summation formula). But for the life of me, I can't figure out how to manipulate that fraction to give the final result.

For example, if I put k=0 into that fraction, I get 0/0, not 5. I tried a bunch of manipulation of the fraction to get sines and cosines, and make the denominator real, but I still can't get a closed form solution that gives the final result.

What am I missing?

Thanks!!

UPDATE: Forgot to mention, this is a discrete time function. k is always an integer.

Last edited by a moderator: May 4, 2017
2. Jul 13, 2009

WolfOfTheSteps

I totally forgot about that factoring trick:

$$1-e^{-j x} = e^{-j x/2}(e^{j x/2}-e^{-j x/2}) = e^{-j x/2}jsin(x/2)$$

That's all I needed!

UPDATE:

Wait, I was wrong.

Even with that factoring trick, I get:

$$e^{-j(\pi k-\pi k/5)}*sin(\pi k)/sin(\pi k/5)$$

But this is still 0/0 for k=0. How do I get the real result?? I'm so frustrated with this!

Last edited: Jul 13, 2009
3. Jul 13, 2009

n!kofeyn

No matter what k is, the top is 0. If k is not 0,+/-5,+/-10,..., then the bottom is not 0, so that is how they get 0 for otherwise. For k=0,+/-5,+/-10,..., you need to find the limit as k approaches those values, because 0/0 has no meaning. Use l'Hospital's rule to evaluate your function at those points.

4. Jul 13, 2009

WolfOfTheSteps

Thanks, that makes sense.

Also, instead of going to L'Hospital's rule, I could just go back to the summation for k=0,+-5, etc and show that it is a summation of ones.... while the fraction would prove the "0 otherwise" for the other k values. This would work too.

Thanks a lot!!

5. Jul 13, 2009

n!kofeyn

No problem. Yea, you're right about the summation giving you the five.

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook