Solving a Recurrence Equation: What Did I Do Wrong?

  • Thread starter Thread starter yoran
  • Start date Start date
  • Tags Tags
    Recurrence
Click For Summary

Homework Help Overview

The discussion revolves around solving a recurrence equation of the form y[k+2] + y[k] = sin(k). Participants are examining the process of finding both the homogeneous and particular solutions, while comparing their findings to a provided answer.

Discussion Character

  • Exploratory, Mathematical reasoning, Problem interpretation

Approaches and Questions Raised

  • The original poster attempts to solve the homogeneous part of the equation and expresses confusion over the role of complex roots in the solution. They also explore finding a particular solution and question the validity of their substitution.

Discussion Status

Participants are actively engaging with the problem, providing feedback on each other's reasoning. Some suggest alternative approaches for finding the particular solution, while others clarify misunderstandings regarding the homogeneous solution. There is an ongoing exploration of the relationship between different forms of the solutions presented.

Contextual Notes

Participants are working under the constraints of a homework assignment, which may limit the information available or the methods they can use. There is a noted confusion regarding the manipulation of trigonometric identities and the implications of complex coefficients in the solutions.

yoran
Messages
116
Reaction score
0
Hi, I'm trying to solve this recurrence equation.

Homework Statement


Solve [tex]y[k+2]+y[k]=sin(k)[/tex].
The answer is already given, it's
[tex]y[k]=c_1sin(\frac{\pi}{2}k) + c_2cos(\frac{\pi}{2}k) + \frac{sin(k)+sin(k-2)}{2(1+cos(2))}[/tex]

Homework Equations




The Attempt at a Solution


First I solve the homogeneous equation.
[tex]y[k+2]+y[k]=0[/tex].
The characteristic function is
[tex]r^2+1=0[/tex]
This has two complex roots [tex]r=i[/tex] and [tex]r=-i[/tex].
Thus the general solution is
[tex]y[k]=c_1(i)^k+c_2(-i)^k[/tex]
If you change the notation for the complex numbers, then it gives
[tex]y[k]=c_1(cos(\frac{\pi}{2}k)+isin(\frac{\pi}{2}k)) + c_2(cos(\frac{-\pi}{2}k)+isin(\frac{-\pi}{2}k))[/tex]
After grouping together it gives
[tex](c_1+c_2)cos(\frac{\pi}{2}k) + (c_1-c_2)isin(\frac{\pi}{2}k)[/tex]
Apparently, the answer is without complex numbers so I guess they only keep the real part. Then it must be so that
[tex]c_1-c_2=0\Leftrightarrow c_1=c_2[/tex]
Then I get as solution
[tex]y[k]=2c_1cos(\frac{\pi}{2}k)[/tex]
which is not the same solution as the general solution in the answer. What did I do wrong?

I also tried to find a particular solution for the non-homogeneous equation. I tried as a solution
[tex]y[k]=\alpha sin(k)[/tex]
with [tex]\alpha[/tex] the parameter to determine. I don't this is the right solution. Can anyone help me with this problem?

Thanks,

Yoran
 
Physics news on Phys.org
They don't "only keep the real part". Plug the solution back into the homogeneous equation; the sine part falls out without imposing any constraint on c1 and c2.
 
Hi,

Thanks for your quick reply. I tried to do what you told me but when I plug it into the homogeneous equation, everything cancels out! I get 0=0. This is what I did:
We have that
[tex]y[k]=(c_1+c_2)cos(\frac{\pi}{2}k)+(c_1-c_2)isin(\frac{\pi}{2}k)[/tex]
I plug it into
[tex]y[k+2]+y[k]=0[/tex]
I get
[tex](c_1+c_2)cos(\frac{\pi}{2}(k+2))+(c_1-c_2)isin(\frac{\pi}{2}(k+2))+(c_1+c_2)cos(\frac{\pi}{2}k)+(c_1-c_2)isin(\frac{\pi}{2}k)=0[/tex]
Because
[tex]cos(\theta + \pi)=-cos(\theta)[/tex] and
[tex]sin(\theta + \pi)=-sin(\theta)[/tex]
I get that
[tex]-(c_1+c_2)cos(\frac{\pi}{2}k)-(c_1-c_2)isin(\frac{\pi}{2}k)+(c_1+c_2)cos(\frac{\pi}{2}k)+(c_1-c_2)isin(\frac{\pi}{2}k)=0[/tex]
which makes
[tex]0=0[/tex]
How come?
 
Actually, you are correct. That was a silly suggestion of me, because 0=0 simply confirms your solution satisfies the equation from which you derived the solution. I will edit this post if I come up with something.

Edit: Wow I had been reading it wrong for a good 45 minutes, sorry! Consider complex coefficients and make the substitution d1=c1+c2 and d2=i(c1-c2).

Can you find the particular solution from here?
 
Last edited:
Hi,

Yeah of course... just give different names as they are just parameters... Thanks a lot.
I tried the substitution [tex]\alpha sin(k)[/tex] because there is [tex]sin(k)[/tex] on the right side of the equation. However, I can't find the right answer. This is what I have done:
Try
[tex]y[k]=\alpha sin(k)[/tex]
Plug it into the equation:
[tex]\alpha sin(k+2) + \alpha sin(k) = sin(k)[/tex]
Then we get
[tex]\alpha = \frac{sin (k)}{2sin(k+1)cos(1)}[/tex]
by using the identity
[tex]sin(\theta) + sin(\alpha) = 2sin(\frac{\theta + \alpha}{2})cos(\frac{\theta - \alpha}{2})[/tex]
The answer for [tex]\alpha[/tex] is obviously not the right answer.
Did I make the wrong substitution?
 
Anyone?
 
yoran said:
Try
[tex]y[k]=\alpha sin(k)[/tex]

Hi yoran! :smile:

Try [tex]y[k]=\alpha\,sin(k\,-\,1)[/tex] . :smile:
 
Ok, I tried
[tex]y[k]=\alpha \sin(k-1)[/tex].
Then I get that
[tex]\alpha = \frac{1}{2\cos 1}[/tex]
So given the solution for the homogeneous equation and this solution, I have that
[tex]y[k]=c_1\sin{\frac{\pi}{2}k} + c_2\cos{\frac{\pi}{2}k} + \frac{1}{2\cos{1}}\sin{k-1}[/tex]
But the answer should be
[tex]y[k]=c_1\sin{\frac{\pi}{2}k} + c_2\cos{\frac{\pi}{2}k} + \frac{\sin{k}+\sin(k-2)}{2(1+\cos{2})}[/tex]
Where does it go wrong in my solution?
 
Hi yoran! :smile:
yoran said:
…I have that
[tex]y[k]=c_1\sin{\frac{\pi}{2}k} + c_2\cos{\frac{\pi}{2}k} + \frac{1}{2\cos{1}}\sin{k-1}[/tex]
But the answer should be
[tex]y[k]=c_1\sin{\frac{\pi}{2}k} + c_2\cos{\frac{\pi}{2}k} + \frac{\sin{k}+\sin(k-2)}{2(1+\cos{2})}[/tex]
Where does it go wrong in my solution?

:smile: But they're the same! :smile:

Standard trig equation: 1 + cos(2) = 2cos²(1).

So [tex]\frac{\sin{k}+\sin(k-2)}{2(1+\cos{2})}[/tex] = … ? :smile:
 
  • #10
Hehe, of course... Thanks a lot!
[tex]\frac{\sin{k}+\sin(k-2)}{2(1+\cos{2})}=\frac{2\sin(k-1)\cos(1)}{4\cos^2(1)}=\frac{\sin(k-1)}{2\cos 1}[/tex]
 

Similar threads

Replies
8
Views
1K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 14 ·
Replies
14
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
12
Views
2K
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
1
Views
2K