Gram-Schmidt Process: Find Basis W for Polynomials P_2

  • Thread starter Thread starter Hiche
  • Start date Start date
  • Tags Tags
    Process
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
2 replies · 2K views
Hiche
Messages
82
Reaction score
0

Homework Statement



The question states that we should use the Gram-Schmidt to find an orthogonal basis for W where W = span {p , q} and p(x) = 1 + x ; q(x) = 1 + 2x^2

Homework Equations



...

The Attempt at a Solution



Let [itex]\{b1, b2\}[/itex] be such a basis. Using the G-S process:
[itex]b1 = p = 1 + x[/itex]
[itex]b2 = q - proj^q _b = q - \frac{<q, b1>} { \|b1\|^2} . b1[/itex]

Is this the right way? The answer I got was [itex]\{b1, b2\}[/itex] = [itex]\{1 + x, 1/2 - x/2 + 2x^2\}[/itex]

Note that p and q belong to the set of polynomials of degrees 2 or less (i.e [itex]\in P_2[/itex]) and [itex]< , >[/itex] denotes an inner product of two components.

Also, presume I need to find further basis -- [itex]b3[/itex] -- is there a generality for [itex]bn[/itex] for example?
 
Last edited:
Physics news on Phys.org
Hiche said:

Homework Statement



The question states that we should use the Gram-Schmidt to find an orthogonal basis for W where W = span {p , q} and p(x) = 1 + x ; q(x) = 1 + 2x^2

Homework Equations



...

The Attempt at a Solution



Let [itex]\{b1, b2\}[/itex] be such a basis. Using the G-S process:
[itex]b1 = p = 1 + x[/itex]
[itex]b2 = q - proj^q _b = q - \frac{<q, b1>} { \|b1\|^2} . b1[/itex]

Is this the right way? The answer I got was [itex]\{b1, b2\}[/itex] = [itex]\{1 + x, 1/2 - x/2 + 2x^2\}[/itex]
I haven't checked your work closely, but this is the right idea. You can check orthogonality by taking the inner product of these two functions. BTW, you didn't say what the inner product was for your space - an integral of some kind?
Hiche said:
Note that p and q belong to the set of polynomials of degrees 2 or less (i.e [itex]\in P_2[/itex]) and [itex]< , >[/itex] denotes an inner product of two components.
Again, what is the inner product you are using?
Hiche said:
Also, presume I need to find further basis -- [itex]b3[/itex] -- is there a generality for [itex]bn[/itex] for example?
{b1, b2} is the basis for W, which is a proper subspace of P2.

The dimension of W is 2, so any basis for W will consist of two lin. independent vectors/functions. This means it is not possible to find another basis function, which is what I think you are asking. If you add a vector or function to a basis, the new addition will necessarily be a linear combination of the other elements in the set.

The dimension of P2 is 3, so a basis for this function space will consist of three lin. independent functions.

If you are asking whether the Gram-Schmidt process can be extended to find more than two orthogonal basis elements - yes, it can. I'm sure you can find an article on wikipedia about this.
 
Oh right. The inner product is defined on [itex]P_2[/itex] as such: [itex]<p, q> = a_0b_0 + a_1b_1 + a_2b_2[/itex] where [itex]a_ns[/itex] are the constants of [itex]p(x)[/itex] and [itex]b_ns[/itex] are the constants of [itex]q(x)[/itex].