Question in title, my progress:

let $z = \cos(x) + i\sin(x)$

then $1 + \cos(x) + \cos(2x) +\dots + \cos(nx) = Re(1 + z + z^2 +\dots + z^n) = Re\left (\dfrac{1-z^{n+1}}{1-z} \right)$

by geometric series;

multiplying $\dfrac{1-z^{n+1}}{1-z}$ by $\overline{1-z}$ we get $1 + \cos(x) + \cos(2x) +\dots + \cos(nx) = Re \left ( \dfrac{(1-z^{n+1})(\overline{1-z})}{|1-z|^2} \right )$

but I am not sure how to proceed from here.

edit: this is for a complex analysis course, so i'd appreciate a hint using complex analysis without using the exponential function

  • $\begingroup$ proofwiki.org/wiki/… OR math.stackexchange.com/questions/117114/… $\endgroup$ – lab bhattacharjee Sep 30 '14 at 16:40
  • $\begingroup$ @labbhattacharjee - Why not vote to close as a duplicate? $\endgroup$ – nbubis Sep 30 '14 at 16:44
  • $\begingroup$ $z=e^{i\theta}$ $\implies Re \left(\dfrac{(1-z^{n+1})(1-\bar z)}{{\lvert 1-z \rvert}^2}\right)=Re \left(\dfrac{(1-e^{i(n+1)\theta})(1-e^{-\theta})}{{\lvert 1-e^{i\theta} \rvert}^2}\right)$. $\endgroup$ – user 170039 Sep 30 '14 at 16:44
  • $\begingroup$ @nbubis, Why don't you start? People may follow suit $\endgroup$ – lab bhattacharjee Sep 30 '14 at 16:48
  • $\begingroup$ @labbhattacharjee - I assumed that if you found the duplicate and didn't vote to close there must have been a good reason. $\endgroup$ – nbubis Sep 30 '14 at 16:49

$\textbf{Hint:}$Use De Moivre's formula to compute $z^{n+1}$.

$\textbf{Edit:}$ The other way to compute this sum is writing $\cos x$ as:

$$\displaystyle \cos x=\frac{e^{ix}+e^{-ix}}{2}$$

In my opinion it's the easiest way. You simply get two geometric series:

$$\sum_{k=0}^{n}\cos kx=\sum_{k=0}^{n} \frac{e^{ikx}+e^{-ikx}}{2}=\frac{1}{2}\left(\sum_{k=0}^{n}e^{ikx}+\sum_{k=0}^{n}e^{-ikx}\right)= \\ = \frac{1}{2}\left(\frac{1-e^{i(n+1)x}}{1-e^{ix}}+\frac{1-e^{-i(n+1)x}}{1-e^{-ix}}\right)$$

It's equal:


Using again formula for $\cos$ you get:

$$\frac{1}{2}\frac{2+2\cos nx -2\cos x -2\cos (n+1)x}{2-2\cos x}$$

  • $\begingroup$ It gets very messy if I switch back to cosines/sines $\endgroup$ – Jonx12 Sep 30 '14 at 16:46
  • $\begingroup$ anymore tips please? $\endgroup$ – Jonx12 Sep 30 '14 at 21:57
  • $\begingroup$ I added one more approach. $\endgroup$ – agha Sep 30 '14 at 22:42

Use $\sin(a) \cos(b) = \frac{1}{2} \sin(b+a) - \frac{1}{2} \sin(b-a)$ and multiply your sum with $\sin\left(x/2\right)$. $$ \sum_{m=0}^{n} \sin\left(\frac{x}{2}\right) \cos(m x) = \frac{1}{2} \sum_{m=0}^n \left\{\sin \left(\left(m+1-\frac{1}{2}\right)x \right) - \sin\left( \left(m-\frac{1}{2}\right)x \right) \right\} $$ The sum telescopes, i.e. $\sum_{m=0}^n \left(f(m+1)-f(m)\right) = f(n+1)-f(0)$, hence $$ \sum_{m=0}^{n} \sin\left(\frac{x}{2}\right) \cos(m x) = \sin \left(\left(n+\frac{1}{2}\right)x \right) - \sin\left(\frac{x}{2} \right) $$ now divide by $\sin\left(\frac{x}{2}\right)$.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.