A problem on finding orthogonal basis and projection

In summary: Once you have an orthogonal basis, you can use the formula to calculate the projection of any vector onto that basis. So in this case, you would plug in the function y = 3(x+x^2) for u and the vectors f_1, f_2, f_3 for the v_i's in the formula you wrote to calculate the projection.
  • #1
visharad
54
0
Use the inner product <f,g> = integral f(x) g(x) dx from 0 to 1 for continuous functions on the inerval [0, 1]

a) Find an orthogonal basis for span = {x, x^2, x^3}

b) Project the function y = 3(x+x^2) onto this basis.
---------------------------------------------------------
I know the following:
Two vectors are orthogonal if their inner product = 0
A set of vectors is orthogonal if <v1,v2> = 0 where v1 and v2 are members of the set and v1 is not equal to v2
If S = {v1, v2, ..., vn} is a basis for inner product space and S is also an orthogonal set, then S is an orthogonal basis.

Regarding projection, I know that if W is a finite dimensional subspace of an inner product space V and W has an orthogonal basis S = {v1, v2, ..., vn} and that u is any vector in V then,
projection of u onto W = <u, v1> v1/||v1||^2 + <u, v2> v2/||v2||^2 + <u, v3> v3/||v3||^2 + ...<u, vn> vn/||vn||^2

I can calculate integrals, but I really do not know how to fit all these together for this problem. I am not sure how to start.
 
Physics news on Phys.org
  • #3
Sorry, but still I did not understand. Can you at least give major steps?
 
  • #4
What are the column vectors of S ?
 
  • #5
Is this correct for part a?

If r + s = n, then
<x^r, x^s> = ∫_0^1 x^n dx = x^(n+1)/(n+1) (0, 1)
= 1^(n+1)/(n+1) - 0^(n+1)/(n+1)
= 1/(n+1)

Let the orthogonal basis = {f1, f2, f3}

Put f1 = 1
f2 = t^2- (<t^2,t>)/(<t,t>)*t = t^2- (1/4)/(1/3)*t = t^2 - 3t/4
f3 = t^3- (<t^3,t>)/(<t,t>)*t - (<t^3,t^2>)/(<t^2,t^2>)*t^2
= t^3 - (1/5)/(1/3)*t - (1/6)/(1/5) * t^2 = t^3-3t/5-(5t^2)/6

The orthogonal basis = {1, t^2 - 3t/4, t^3-3t/5-(5t^2)/6}

--------------
How do I solve part b ?
 
  • #6
visharad said:
Let the orthogonal basis = {f1, f2, f3}

Put f1 = 1
Shouldn't that be [itex] f_1 = t [/itex] since you're orthogonalizing the basis [itex] \{x, x^2, x^3\}[/itex]? You should set [itex] f_1[/itex] equal to the first vector of the original basis. Actually, it looks like you took [itex] f_1 = t [/itex] for the calculation of [itex] f_2 [/itex]...
f2 = t^2- (<t^2,t>)/(<t,t>)*t = t^2- (1/4)/(1/3)*t = t^2 - 3t/4
f3 = t^3- (<t^3,t>)/(<t,t>)*t - (<t^3,t^2>)/(<t^2,t^2>)*t^2
= t^3 - (1/5)/(1/3)*t - (1/6)/(1/5) * t^2 = t^3-3t/5-(5t^2)/6
Make sure you're following the formula I linked. You should be taking the projection along the orthogonal vectors you calculated previously, so you should have
[tex]
f_3 = t^3 - \frac{\langle t^3, f_1 \rangle}{\|f_1\|^2} f_1 - \frac{\langle t^3, f_2 \rangle}{\|f_2\|^2} f_2
[/tex]

I think you calculated [itex] f_2[/itex] correctly, but not [itex] f_3 [/itex].
How do I solve part b ?
You wrote the formula for this in your first post. Once you finish part (a), you can use it.
 

FAQ: A problem on finding orthogonal basis and projection

1. What is an orthogonal basis?

An orthogonal basis is a set of vectors that are mutually perpendicular to each other and have unit length. This means that the inner product (or dot product) of any two vectors in the basis is equal to 0, and the length of each vector is equal to 1.

2. Why is finding an orthogonal basis important?

Finding an orthogonal basis is important because it allows us to express any vector in a space as a linear combination of these basis vectors. This makes it easier to solve problems involving vector spaces, such as finding projections and solving systems of linear equations.

3. How do you find an orthogonal basis?

There are several methods for finding an orthogonal basis, including Gram-Schmidt process, QR decomposition, and singular value decomposition. These methods involve manipulating the given vectors in a specific way to create a set of orthogonal vectors.

4. What is the projection of a vector onto a subspace?

The projection of a vector onto a subspace is the closest vector to the original vector in the subspace. It is calculated by finding the orthogonal projection of the vector onto each vector in the orthogonal basis of the subspace and adding them together.

5. How is finding an orthogonal basis related to solving systems of linear equations?

By finding an orthogonal basis for the vector space spanned by the coefficients of the equations, we can rewrite the system of equations in terms of this basis. This makes it easier to solve the system, as we can solve for each basis vector separately and then combine the solutions to obtain the solution for the original system.

Back
Top