Show that set is an orthonormal set in PC(0,l)

  • Context: Graduate 
  • Thread starter Thread starter Shackleford
  • Start date Start date
  • Tags Tags
    Set
Click For Summary
SUMMARY

The set \{ (2/l)^{1/2} sin(n-1/2)(\pi x/l)\}^\infty_1 is confirmed as an orthonormal set in PC(0,l) by demonstrating that the inner product <\phi_m, \phi_n> equals the Kronecker delta δ_{mn}. The evaluation of the integral <\phi_m, \phi_n> = (2/l) ∫_0^l sin(m-1/2)(\pi x/l) sin(n-1/2)(\pi x/l) dx shows that it equals 1 for m=n and 0 for m≠n. The discussion highlights the importance of using the correct trigonometric identities and integral evaluations to avoid errors in the derivation.

PREREQUISITES
  • Understanding of orthonormal sets in functional analysis
  • Familiarity with trigonometric identities, specifically 2 sin θ sin ψ = cos(θ - ψ) - cos(θ + ψ)
  • Knowledge of integral calculus, particularly definite integrals involving sine functions
  • Experience with inner product spaces and the concept of the Kronecker delta
NEXT STEPS
  • Study the properties of orthonormal sets in Hilbert spaces
  • Learn about the application of trigonometric identities in integral evaluations
  • Explore the derivation of Fourier series and their orthonormal properties
  • Investigate common pitfalls in evaluating integrals involving sine and cosine functions
USEFUL FOR

Mathematicians, physicists, and students studying functional analysis or Fourier analysis, particularly those interested in orthonormal sets and their applications in various fields.

Shackleford
Messages
1,649
Reaction score
2
Show that set \{ (2/l)^{1/2} sin(n-1/2)(\pi x/l)\}^\infty_1 is an orthonormal set in PC(0,l).
Of course, I need to show that &lt;\phi_m, \phi_n&gt; = \delta_{mn}
&lt;\phi_m, \phi_n&gt; = (2/l) \int_0^l sin(m-1/2)(\pi x/l) sin(n-1/2)(\pi x/l)\, dx<br /> \\ = 1/l \int_0^l cos(m-n)(\pi x/l) cos(m+n+1)(\pi x/l)\, dx<br /> \\ = \left( \frac 1 l \right) \frac {l} {\pi(m-n)} sin(m-n)(\pi x/l) \Big|_0^l - \left( \frac 1 l \right) \frac {l} {\pi(m-n)} sin(m+n+1)(\pi x/l)\ \Big|_0^l<br /> \\ = \frac {1} {\pi(m-n)} sin(m-n)(\pi x/l) \Big|_0^l - \frac {1} {\pi(m-n)} sin(m+n+1)(\pi x/l)\ \Big|_0^l<br />

I used the trig identity:
2 sin \theta sin \psi = cos(\theta - \psi) - cos(\theta + \psi)
 
Physics news on Phys.org
What is your question? The expression you have (you should check it out) = 0 for ##m\ne n## and =1 for ##m=n##.
 
mathman said:
What is your question? The expression you have (you should check it out) = 0 for ##m\ne n## and =1 for ##m=n##.

Thanks for the reply. When I looked at it, I could swear it's zero in both cases. Let me check again.
 
<br /> \\ = \frac {1} {\pi(m-n)} sin(m-n)(\pi x/l) \Big|_0^l - \frac {1} {\pi(m-n)} sin(m+n+1)(\pi x/l)\ \Big|_0^l<br />

When you evaluate at x = l and x = 0, the argument for sine is either zero or a multiple of pi. When m = n, there's an indeterminate form in the common term.
 
When m=n, I suggest you don't use the cos identity, but use the original form, where the integral is ##2/l\int_0^lsin^2((m-1/2)(\pi x/l))dx##.

I think you have some errors in the derivation. You have the product of cos terms, rather than the difference, although this seems to have been corrected for the next line. When you integrated the cos (m+n+!)...) term, you divided by m-n.
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
5
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
Replies
28
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K