The Student Room Group

Summation of trigonometric identities

Prove that k=0nsinkθ=cos(θ/2)cos(n+1/2)θ2sin(θ/2)\displaystyle\sum_{k=0}^n \sin k\theta = \dfrac{\cos (\theta /2) - cos (n + 1/2)\theta}{2 \sin (\theta /2)}

Attempt:
k=0nsinkθ=sin0+sinθ+sin2θ+...+sin(nθ)[br][br]by de moivre’s theorem,k=0nei(kθ)=0+ei(θ)+ei(2θ)+ei(3θ)+...+ei(nθ)[br][br]by sum of G.P.,[br][br]=ei(θ)(1ei(nθ))1ei(θ)[br][br]=ei(θ)(1ei(nθ))(1+ei(θ)(1ei(θ))(1+ei(θ))[br]\displaystyle\sum_{k=0}^n \sin k\theta = \sin 0 + \sin \theta + \sin 2\theta + ... + \sin (n\theta)[br][br]\text{by de moivre's theorem,} \displaystyle\sum_{k=0}^n e^{i(k\theta)} = 0 + e^{i(\theta)} + e^{i(2\theta)} +e^{i(3\theta)} + ... + e^{i(n\theta)}[br][br]\text{by sum of G.P.,} [br][br]= \dfrac{e^{i(\theta)}(1 - e^{i(n\theta)})}{1- e^{i(\theta)}}[br][br]= \dfrac{e^{i(\theta)}(1 - e^{i(n\theta)})(1 + e^{i(\theta)}}{(1- e^{i(\theta)})(1+ e^{i(\theta)})}[br]

But the powers of the exp don't match the target of which I'm supposed to prove. Pointers anyone?
Reply 1
Original post by johnconnor92
Prove that k=0nsinkθ=cos(θ/2)cos(n+1/2)θ2sin(θ/2)\displaystyle\sum_{k=0}^n \sin k\theta = \dfrac{\cos (\theta /2) - cos (n + 1/2)\theta}{2 \sin (\theta /2)}

Attempt:
k=0nsinkθ=sin0+sinθ+sin2θ+...+sin(nθ)[br][br]by de moivre’s theorem,k=0nei(kθ)=0+ei(θ)+ei(2θ)+ei(3θ)+...+ei(nθ)[br][br]by sum of G.P.,[br][br]=ei(θ)(1ei(nθ))1ei(θ)[br][br]=ei(θ)(1ei(nθ))(1+ei(θ)(1ei(θ))(1+ei(θ))[br]\displaystyle\sum_{k=0}^n \sin k\theta = \sin 0 + \sin \theta + \sin 2\theta + ... + \sin (n\theta)[br][br]\text{by de moivre's theorem,} \displaystyle\sum_{k=0}^n e^{i(k\theta)} = 0 + e^{i(\theta)} + e^{i(2\theta)} +e^{i(3\theta)} + ... + e^{i(n\theta)}[br][br]\text{by sum of G.P.,} [br][br]= \dfrac{e^{i(\theta)}(1 - e^{i(n\theta)})}{1- e^{i(\theta)}}[br][br]= \dfrac{e^{i(\theta)}(1 - e^{i(n\theta)})(1 + e^{i(\theta)}}{(1- e^{i(\theta)})(1+ e^{i(\theta)})}[br]

But the powers of the exp don't match the target of which I'm supposed to prove. Pointers anyone?


e^0=1
Reply 2
Gah stupid mistake. Here's a revised one.

Prove that k=0nsinkθ=cos(θ/2)cos(n+1/2)θ2sin(θ/2)\displaystyle\sum_{k=0}^n \sin k\theta = \dfrac{\cos (\theta /2) - cos (n + 1/2)\theta}{2 \sin (\theta /2)}

Attempt:
k=0nsinkθ=sin0+sinθ+sin2θ+...+sin(nθ)[br][br]by de moivre’s theorem,k=0nei(kθ)=1+ei(θ)+ei(2θ)+ei(3θ)+...+ei(nθ)[br][br]by sum of G.P.,[br][br]=(1ei((n+1)θ))1ei(θ)[br][br]=(1ei((n+1)θ))(1+ei(θ)(1ei(θ))(1+ei(θ))[br]\displaystyle\sum_{k=0}^n \sin k\theta = \sin 0 + \sin \theta + \sin 2\theta + ... + \sin (n\theta)[br][br]\text{by de moivre's theorem,} \displaystyle\sum_{k=0}^n e^{i(k\theta)} = 1 + e^{i(\theta)} + e^{i(2\theta)} +e^{i(3\theta)} + ... + e^{i(n\theta)}[br][br]\text{by sum of G.P.,} [br][br]= \dfrac{(1 - e^{i((n+1)\theta)})}{1- e^{i(\theta)}}[br][br]= \dfrac{(1 - e^{i((n+1)\theta)})(1 + e^{i(\theta)}}{(1- e^{i(\theta)})(1+ e^{i(\theta)})}[br]

But the powers of the exp don't match the target of which I'm supposed to prove. Pointers anyone?
Reply 3
Don't they? I've not looked too closely but I think you may have forgotten that e^(ix) = cosx + isinx, so you've summed cos(kx) + i sin(kx), so you ought to be taking just the imaginary part of that sum you've got, that will probably mess about with it suitably well as to make it look similar.
Reply 4
You could start like this:

k=0nsinkθ=12sin12θk=0n2sin(kθ)sin(12θ)\sum_{k=0}^n \sin k\theta=\frac{1}{2\sin \frac{1}{2}\theta}\sum_{k=0}^n 2\sin(k\theta)\sin(\frac{1}{2} \theta)
Reply 5
Original post by Allofthem
Don't they? I've not looked too closely but I think you may have forgotten that e^(ix) = cosx + isinx, so you've summed cos(kx) + i sin(kx), so you ought to be taking just the imaginary part of that sum you've got, that will probably mess about with it suitably well as to make it look similar.


I do know about that, but if we look at the powers of the natural constant how are we supposed to get a half out of the terms? I didn't go on because I hesitated on seeing the powers.

Original post by BabyMaths
You could start like this:

k=0nsinkθ=12sin12θk=0n2sin(kθ)sin(12θ)\sum_{k=0}^n \sin k\theta=\frac{1}{2\sin \frac{1}{2}\theta}\sum_{k=0}^n 2\sin(k\theta)\sin(\frac{1}{2} \theta)


I know about this method. Tried it and done it, too. But I just want to nail the question with the method here.
Reply 6
Original post by johnconnor92
= \dfrac{(1 - e^{i((n+1)\theta)})}{1- e^{i(\theta)}}As you probably know, You want to take the imaginary part of this, which means you need to make the denominator easier to deal with.

The normal way of doing this would be to multiply top and bottom by (1eiθ)(1-e^{-i\theta}), which makes the denominator real.

A better method here is to multiply top and bottom by eiθ/2e^{-i\theta/2}, which leaves the denominator as eiθ/2eiθ/2e^{-i\theta/2}-e^{i\theta/2}, which is pure imaginary. It should work out fairly easily from there.
Reply 7
Original post by DFranklin
As you probably know, You want to take the imaginary part of this, which means you need to make the denominator easier to deal with.

The normal way of doing this would be to multiply top and bottom by (1eiθ)(1-e^{-i\theta}), which makes the denominator real.

A better method here is to multiply top and bottom by eiθ/2e^{-i\theta/2}, which leaves the denominator as eiθ/2eiθ/2e^{-i\theta/2}-e^{i\theta/2}, which is pure imaginary. It should work out fairly easily from there.


OF COURSE! I got the complex conjugate for exponentials wrong! OMGWTFBBQ
Reply 8
Original post by johnconnor92
OF COURSE! I got the complex conjugate for exponentials wrong! OMGWTFBBQ
Note what I said about the "better" method here (which doesn't use the complex conjugate). It's a trick worth knowing.
Reply 9
Original post by DFranklin
As you probably know, You want to take the imaginary part of this, which means you need to make the denominator easier to deal with. The normal way of doing this would be to multiply top and bottom by (1eiθ)(1-e^{-i\theta}), which makes the denominator real.


multiplying with complex conjugate,(1ei((n+1)θ))(1ei(θ)(1ei(θ))(1ei(θ))[br][br]=1eiθei(n+1)θ+einθ1eiθeiθ+1[br][br]taking the imaginary parts and solving the denominator,[br]=1(cosθ+isinθ)+(cosnθ+isinnθ)(cos(n+1)θ+isin(n+1)θ)22cosθ[br][br]=sinθ+sinnθsin(n+1)θ)4sin2(θ/2)[br][br]=2cos(θ/2)sin(θ/2)2sin(θ/2)2sin(θ/2)cos(n+1/2)θ4sin2(θ/2)[br][br]=cos(θ/2)+cos(n+1/2)θ2sin(θ/2)\text{multiplying with complex conjugate,} \dfrac{(1 - e^{i((n+1)\theta)})(1 - e^{-i(\theta)}}{(1- e^{i(\theta)})(1- e^{-i(\theta)})}[br][br]= \dfrac{1-e^{-i\theta}-e^{i(n+1)\theta}+e^{in\theta}}{1-e^{-i\theta} - e^{i\theta} +1 }[br][br]\text{taking the imaginary parts and solving the denominator,}[br]= \dfrac{1-(\cos\theta +i\sin\theta)+(\cos n\theta + i\sin n\theta) - (\cos (n+1)\theta + i\sin (n+1)\theta)}{2-2\cos \theta}[br][br]= \dfrac{-\sin\theta+ \sin n\theta -\sin (n+1)\theta)}{4\sin^2(\theta/2)}[br][br]= \dfrac{-2\cos(\theta/2)\sin(\theta/2)-2\sin(\theta/2) - 2\sin(\theta/2)\cos(n+1/2)\theta}{4\sin^2(\theta/2)}[br][br]= \dfrac{\cos(\theta/2)+ \cos(n+1/2)\theta}{2\sin(\theta/2)}
Hmm... Something went wrong.

A better method here is to multiply top and bottom by eiθ/2e^{-i\theta/2}, which leaves the denominator as eiθ/2eiθ/2e^{-i\theta/2}-e^{i\theta/2}, which is pure imaginary. It should work out fairly easily from there.

(1ei((n+1)θ))(eiθ/2)(1ei(θ))(eiθ/2)[br][br]=eiθ/2ei[(n+1)1/2]θeiθ/2eiθ/2[br][br]=cos(θ/2)isin(θ/2)cos(n+1/2)θ+sin(n+1/2)θ)cos(θ/2)isin(θ/2)isin(θ/2)cos(θ/2)[br][br]taking the imaginary parts,[br][br]=sin(θ/2)+sin(n+1/2)θ2sin(θ/2)\dfrac{(1 - e^{i((n+1)\theta)})(e^{-i\theta/2})}{(1- e^{i(\theta)})(e^{-i\theta/2})}[br][br]= \dfrac{e^{-i\theta/2} - e^{i[(n+1)-1/2]\theta}}{e^{-i\theta/2} - e^{i\theta/2}}[br][br]= \dfrac{\cos(\theta/2) - i\sin(\theta/2) - \cos(n+1/2)\theta +\sin(n+1/2)\theta)}{cos(\theta/2) - i\sin(\theta/2) - i\sin(\theta/2) -\cos(\theta/2)}[br][br]\text{taking the imaginary parts,}[br][br]= \dfrac{-\sin(\theta/2)+\sin(n+1/2)\theta}{-2\sin(\theta/2)}

This is amazing! I always thought denominators with an imaginary numbers cannot be navigated like that of real numbers. But what went wrong in the above calculation? Thank you so much!
(edited 11 years ago)
Reply 10
Original post by DFranklin
Note what I said about the "better" method here (which doesn't use the complex conjugate). It's a trick worth knowing.


I proved the summation via the complex conjugate method. Here's another attempt at the pointer you gave me:

(1ei((n+1)θ))(eiθ/2)(1ei(θ))(eiθ/2)[br]=eiθ/2ei[(n+1)1/2]θeiθ/2eiθ/2[br]=cos(θ/2)isin(θ/2)cos(n+1/2)θsin(n+1/2)θ)cos(θ/2)isin(θ/2)isin(θ/2)cos(θ/2)\dfrac{(1 - e^{i((n+1)\theta)})(e^{-i\theta/2})}{(1- e^{i(\theta)})(e^{-i\theta/2})}[br]= \dfrac{e^{-i\theta/2} - e^{i[(n+1)-1/2]\theta}}{e^{-i\theta/2} - e^{i\theta/2}}[br]= \dfrac{\cos(\theta/2) - i\sin(\theta/2) - \cos(n+1/2)\theta -\sin(n+1/2)\theta)}{cos(\theta/2) - i\sin(\theta/2) - i\sin(\theta/2) -\cos(\theta/2)}

I don't think I made any mistakes with the working above (god forbid), but what I really don't know is whether I should be taking the real/imaginary part of the expression. The real part give the answer at the numerator, but what about the denominator?

Unparseable latex formula:

\text{taking the real/imaginary part (?)}[br]= \dfrac{\cos(\theta/2)-\cos(n+1/2)\theta}{-2i\sin(\theta/2)}}



Please help. Thank you!
You need to take the imaginary part of the whole expression. (You can multiply top and bottom by i to make the denominator real).
Reply 12
Original post by DFranklin
You need to take the imaginary part of the whole expression. (You can multiply top and bottom by i to make the denominator real).


What? So after the multiplication of the e^(-i theta/2) factor i STILL have to multiply another i factor into the expression? Where can I learn more about this? Thank you!

On a side note, how does a fraction with an imaginary denominator differ from that of a real denominator? Are there any significant changes in its behaviour when the numerator/denominator changes by 1 or 2?
Reply 13
Original post by DFranklin
You need to take the imaginary part of the whole expression. (You can multiply top and bottom by i to make the denominator real).


Original post by johnconnor92
What? So after the multiplication of the e^(-i theta/2) factor i STILL have to multiply another i factor into the expression? Where can I learn more about this? Thank you!

On a side note, how does a fraction with an imaginary denominator differ from that of a real denominator? Are there any significant changes in its behaviour when the numerator/denominator changes by 1 or 2?


Anyone? Thanks in advance.
Reply 14
Original post by johnconnor92
On a side note, how does a fraction with an imaginary denominator differ from that of a real denominator? Are there any significant changes in its behaviour when the numerator/denominator changes by 1 or 2?


Hm? Rationalizing the denominator doesn't change the number, it's just that you cannot identify the real/imaginary part before you rationalize the denominator.

For example, take the complex number z:

[br]z=a+bic+di[br]Im(z)bd[br][br]z=\dfrac{a+bi}{c+di}[br]Im(z)\not=\dfrac{b}{d}[br]

Rather:

[br]z=a+bic+di=ac+bdadi+bcic2+d2[br]Im(z)=ad+bcc2+d2[br]z=\dfrac{a+bi}{c+di}=\dfrac{ac+bd-adi+bci}{c^2+d^2}[br]Im(z)=\dfrac{-ad+bc}{c^2+d^2}

As always, sorry if I misunderstood your question.
Reply 15
Original post by aznkid66
Hm? Rationalizing the denominator doesn't change the number, it's just that you cannot identify the real/imaginary part before you rationalize the denominator.

For example, take the complex number z:

[br]z=a+bic+di[br]Im(z)bd[br][br]z=\dfrac{a+bi}{c+di}[br]Im(z)\not=\dfrac{b}{d}[br]

Rather:

[br]z=a+bic+di=ac+bdadi+bcic2+d2[br]Im(z)=ad+bcc2+d2[br]z=\dfrac{a+bi}{c+di}=\dfrac{ac+bd-adi+bci}{c^2+d^2}[br]Im(z)=\dfrac{-ad+bc}{c^2+d^2}

As always, sorry if I misunderstood your question.


That's exactly what I was confused about! DFranklin asked a factor of e^{i theta/2), and doing so produces a pure imaginary denominator which, surprisingly, gives the required answer in imaginary form. But why?
Original post by johnconnor92
That's exactly what I was confused about! DFranklin asked a factor of e^{i theta/2), and doing so produces a pure imaginary denominator which, surprisingly, gives the required answer in imaginary form. But why?
I'm really not sure what your problem is, which is why I haven't responded..

If we want to find the imaginary part of a+ibid\dfrac{a+ib}{id}, we can multiply top and buttom by i: aibd\dfrac{ai-b}{-d} and then read off the imaginary part (a/-d) = -a/d.
Reply 17
Original post by DFranklin
I'm really not sure what your problem is, which is why I haven't responded..

If we want to find the imaginary part of a+ibid\dfrac{a+ib}{id}, we can multiply top and buttom by i: aibd\dfrac{ai-b}{-d} and then read off the imaginary part (a/-d) = -a/d.


Oh so now i see your trick you hinted. This is amazing! Thank you so very much for teaching me this! :yes:

Quick Reply

Latest