Orthonormal Sets - Find a projection matrix - Linear Algebra

In summary, the task is to find the projection matrix P that projects vectors in R4 onto R(A), where A is a 4x2 matrix. To do this, we first need to determine R(A), which we can do by multiplying A with a general vector v=[x, y] and simplifying the result. This gives us a basis for R(A). Next, we need to use the formula projSx = (x * u)u, where S is a vector subspace and x is a vector, to find the projection matrix. In this case, we can define x as the standard basis for R4 and use AATx to find the projection matrix P.
  • #1
aargoo
3
0

Homework Statement


Let A be the 4x2 matrix
|1/2 -1/2|
|1/2 -1/2|
|1/2 1/2|
|1/2 1/2|

Find the projection matrix P that projects vectors in R4 onto R(A)

Homework Equations


projSx = (x * u)u where S is a vector subspace and x is a vector

The Attempt at a Solution


v1 = (1/2, 1/2, 1/2, 1/2)T
v2 = (-1/2, -1/2, 1/2, 1/2)T
v1v2 = 0, hence the vectors are orthogonal
||v1|| = 1
||v2|| = 1, hence they form an orthonormal basis for R2

R(A) = span{(1/2, -1/2)T, (1/2, -1/2)T, (1/2, 1/2)T, (1/2, 1/2)T} = span{(1/2, -1/2)T, (1/2, 1/2)T}

And from here I am a bit lost. Would I define x as the standard basis for R4 and find projR(A)x ?

The answer from the book is given as
[.5 -.5 0 0]
[-.5 .5 0 0]
[0 0 .5 -.5]
[0 0 -.5 .5]

Thanks

Note: This is from a section before the Gram-Schmidt orthogonalization process
 
Last edited:
Physics news on Phys.org
  • #2
The first thing I would do is determine R(A). Given any vector v= [x, y], we have
[tex]Av= \begin{bmatrix}\frac{1}{2} & -\frac{1}{2} \\ \frac{1}{2} & -\frac{1}{2} \\ \frac{1}{2} & \frac{1}{2} \\ \frac{1}{2} & \frac{1}{2}\end{bmatrix}\begin{bmatrix}x \\ y \end{bmatrix}= \begin{bmatrix}\frac{1}{2}x- \frac{1}{2}y \\ \frac{1}{2}x- \frac{1}{2}y \\ \frac{1}{2}x+ \frac{1}{2}y \\ \frac{1}{2}x+ \frac{1}{2}y\end{bmatrix}[/tex].

We can write that result as
[tex]\begin{bmatrix}a \\ a \\ b \\ b\end{bmatrix}= \begin{bmatrix}a \\ a \\ 0 \\ 0 \end{bmatrix}+ \begin{bmatrix}0 \\ 0 \\ b \\ b \end{bmatrix}= a\begin{bmatrix}1 \\ 1 \\ 0 \\ 0 \end{bmatrix}+ b\begin{bmatrix}0 \\ 0 \\ 1 \\ 1 \end{bmatrix}[/tex]
where a= x/2+ y/2 and b= x/2- y/2. That is, a basis for R(A) is [tex]\{\begin{bmatrix}1 \\ 1 \\ 0 \\ 0 \end{bmatrix}, \begin{bmatrix} 0 \\ 0 \\ 1 \\ 1\end{bmatrix}\}[/tex].
 
  • #3
Hmm, I'm still trying to figure out if I'm missing a formula somewhere...
A =
[1 0
1 0
0 1
0 1]

So, projAx = AATx
=
[1 1 0 0
1 1 0 0
0 0 1 1
0 0 1 1] x

I'm now confused as to how to find x? It should be a 4x4, correct?
 

FAQ: Orthonormal Sets - Find a projection matrix - Linear Algebra

What is an orthonormal set in linear algebra?

An orthonormal set in linear algebra is a set of vectors that are all orthogonal (perpendicular) to each other and have a unit length. This means that the dot product of any two vectors in the set is 0 and the magnitude of each vector is 1.

What is the purpose of finding a projection matrix in linear algebra?

The purpose of finding a projection matrix in linear algebra is to represent a vector in a new coordinate system. This is useful for solving systems of equations, finding the best approximation of a vector, and understanding the relationship between different vectors.

How do you find a projection matrix for an orthonormal set?

To find a projection matrix for an orthonormal set, you can use the formula P = A(A^T * A)^-1 * A^T, where A is the matrix containing the orthonormal vectors. This will give you a square matrix that, when multiplied by a vector, will project the vector onto the subspace spanned by the orthonormal set.

Can a projection matrix be used for non-orthonormal sets?

Yes, a projection matrix can be used for non-orthonormal sets. However, the matrix will not have some of the nice properties that come with an orthonormal set, such as being symmetric and orthogonal. It is still possible to use the formula mentioned in question 3, but the resulting matrix may not be as intuitive to interpret.

What are some real-world applications of projection matrices?

Projection matrices have many real-world applications, including image and signal processing, data compression, computer graphics, and machine learning. They are also used in various engineering fields, such as control systems and robotics, for modeling and analyzing physical systems.

Similar threads

Replies
2
Views
2K
Replies
2
Views
688
Replies
4
Views
821
Replies
8
Views
2K
Replies
3
Views
2K
Back
Top