Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Verifying Bessel's Differential Equation

  1. Jul 26, 2010 #1
    Hey everyone, I am trying to use the Bessel function:

    [tex]f(x)= \sum_{k=0}^\infty \frac{(-1)^{k}x^{2k+1}}{2^{2k+1}(k!)(k+1)!}[/tex]

    To verify the Differential Equation:


    So I thought this was gonna be really easy, but I am having some difficulties. First I calculated y,y',y'' to be:

    [tex]y= \sum_{k=0}^\infty \frac{(-1)^{k}x^{2k+1}}{2^{2k+1}(k!)(k+1)!}[/tex]

    [tex]y'= \sum_{k=0}^\infty \frac{(-1)^{k}(2k+1)x^{2k}}{2^{2k+1}(k!)(k+1)!}[/tex]

    [tex]y''= \sum_{k=0}^\infty \frac{(-1)^{k}(2k+1)(2k)x^{2k-1}}{2^{2k+1}(k!)(k+1)!}[/tex]

    And then I just pluged those into the differential equation to yield:

    [tex]x^2 \sum_{k=0}^\infty \frac{(-1)^{k}(2k+1)(2k)x^{2k-1}}{2^{2k+1}(k!)(k+1)!} + x\sum_{k=0}^\infty \frac{(-1)^{k}(2k+1)x^{2k}}{2^{2k+1}(k!)(k+1)!} + (x^2-1)\sum_{k=0}^\infty \frac{(-1)^{k}x^{2k+1}}{2^{2k+1}(k!)(k+1)!} = 0 [/tex]

    Then I just put the x terms in the summation since they are constant to get:

    [tex]\sum_{k=0}^\infty \frac{(-1)^{k}(2k+1)(2k)x^{2k+1}}{2^{2k+1}(k!)(k+1)!} + \sum_{k=0}^\infty \frac{(-1)^{k}(2k+1)x^{2k+1}}{2^{2k+1}(k!)(k+1)!} + \sum_{k=0}^\infty \frac{(-1)^{k}(x^2-1)x^{2k+1}}{2^{2k+1}(k!)(k+1)!} = 0 [/tex]

    And since the summations are all indexed the same and all have the same denominator, I combined them to one summation with one fraction:

    [tex]\sum_{k=0}^\infty \frac{(-1)^{k}(2k+1)(2k)x^{2k+1} + (-1)^{k}(2k+1)x^{2k+1} + (-1)^{k}(x^2-1)x^{2k+1}}{2^{2k+1}(k!)(k+1)!} = 0[/tex]

    But now that I have one huge fraction, after all the simplification I have done, I can't get the numerator to zero out since the DE is homogeneous.

    Did I make an error in my calculation? Have I done everything correct, but there is even MORE simplification? And if so, can you help me? Or am I just approaching this problem completely wrong?

    Thanks a TON!

  2. jcsd
  3. Jul 26, 2010 #2
    You should have expanded the term:

    [tex]\left(x^2-1\right)\sum _{k=0}^{\infty } \frac{(-1)^kx^{2k+1}}{2^{2k+1}k!(k+1)!}[/tex]

    into 2 terms:

    [tex]x^2\sum _{k=0}^{\infty } \frac{(-1)^kx^{2k+1}}{2^{2k+1}k!(k+1)!}-\sum _{k=0}^{\infty } \frac{(-1)^kx^{2k+1}}{2^{2k+1}k!(k+1)!}[/tex]

    You shift the index, and combine it again.
  4. Jul 26, 2010 #3
    I did that at first, but I thought that would'd help at all. Expanding that, putting the x's in the summation and then combining, I got:

    [tex]\sum_{k=0}^\infty \frac{(-1)^k(2k+1)(2k)x^{2k+1}+(-1)^k(2k+1)x^{2k+1}+(-1)^kx^{2k+3}-(-1)^kx^{2k+1}}{2^{2k+1}(k!)(k+1)!}[/tex]

    But that still looks like it is going no where. The numerator doesn't look like it is zero, unless it is really cryptic and I am not catching it. Help?
  5. Jul 27, 2010 #4
    You didn't do the shifting of the summation variables. You should do the following:

    [tex]\sum _{k=0}^{\infty } \frac{(-1)^k(2k+1)(2k)x^{2k+1}+(-1)^k(2k+1)x^{2k+1}+(-1)^kx^{2k+3}-(-1)^kx^{2k+1}}{2^{2k+1}k!(k+1)!}[/tex]

    [tex]\sum _{k=0}^{\infty } \frac{(-1)^k(2k+1)(2k)x^{2k+1}+(-1)^k(2k+1)x^{2k+1}-(-1)^kx^{2k+1}}{2^{2k+1}k!(k+1)!}+\sum _{k=0}^{\infty } \frac{(-1)^kx^{2k+3}}{2^{2k+1}k!(k+1)!}[/tex]

    Do the shifting in this step:

    [tex]\sum _{k=0}^{\infty } \frac{(-1)^k(2k+1)(2k)x^{2k+1}+(-1)^k(2k+1)x^{2k+1}-(-1)^kx^{2k+1}}{2^{2k+1}k!(k+1)!}+\sum _{k=1}^{\infty } \frac{(-1)^{k-1}x^{2k+1}}{2^{2k-1}(k-1)!k!}[/tex]

    Grouping it together again:

    [tex]\sum _{k=1}^{\infty } \frac{(-1)^k(2k+1)(2k)x^{2k+1}+(-1)^k(2k+1)x^{2k+1}-(-1)^kx^{2k+1}+4k(k+1)(-1)^{k-1}x^{2k+1}}{2^{2k+1}k!(k+1)!}[/tex]

    You should be able to show that it is indeed zero.
  6. Jul 27, 2010 #5
    Ah, I see. That is indeed zero. Thanks a MILLION! However, I am a little fuzzy how you went from your 3rd to 4th step in your last post. I understand that you let j=k-1 to shift the end summation, but how were you able to recombine it back with the first summation when that one was indexed at k=0? I don't know if this is a stupid question, but that transition from step 3 to 4 just doesn't look natural to me.
  7. Jul 27, 2010 #6
    Since I take out the k=0 term, and it is again zero! I just skip a bit.
  8. Jul 27, 2010 #7
    AHH! I see. Duh. Thanks soooooo much! :D
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook