91Ó°ÊÓ

In the matrix diagonalization process, a key step is finding eigenvalues and their corresponding eigenvectors. These elements are essential in transforming a matrix into its simpler form, allowing easier analysis and computations. To start, eigenvalues are found by solving the characteristic equation, usually taking the form of \( \det(A - \lambda I) = 0 \). Here, \( A \) is the given matrix, \( \lambda \) represents the unknown eigenvalues, and \( I \) is the identity matrix. For the matrix in question:

• The characteristic equation simplifies to \( (a - \lambda)^2 = b^2 \), resulting from the determinant of the matrix \( A - \lambda I \).
• This leads to the discovery of two eigenvalues: \( \lambda_1 = a + b \) and \( \lambda_2 = a - b \).

Eigenvectors are vectors that indicate the direction of "stretching" without changing direction, when a linear transformation is applied. Once eigenvalues are found, eigenvectors can be calculated by solving \( (A - \lambda I) \mathbf{v} = 0 \), yielding vectors for each eigenvalue:

• For \( \lambda_1 = a + b \), a valid eigenvector is \( \mathbf{v_1} = \begin{bmatrix} 1 \ 1 \end{bmatrix} \).
• For \( \lambda_2 = a - b \), a valid eigenvector is \( \mathbf{v_2} = \begin{bmatrix} 1 \ -1 \end{bmatrix} \).

Orthogonalization is the process of converting a set of vectors into mutually orthogonal vectors, meaning they are perpendicular to each other in space. This is a critical step in orthogonal diagonalization, ensuring the resulting matrix \( P \) is an orthogonal matrix.

For each eigenvector obtained, normalizing is performed to achieve this:

• Normalization involves adjusting the length of a vector to 1 to facilitate orthogonality. This is done by dividing each component of the vector by its length.
• For example, for \( \mathbf{v_1} = \begin{bmatrix} 1 \ 1 \end{bmatrix} \), the normalized orthogonal eigenvector is \( \mathbf{u_1} = \frac{1}{\sqrt{2}} \begin{bmatrix} 1 \ 1 \end{bmatrix} \).
• Similarly, \( \mathbf{v_2} = \begin{bmatrix} 1 \ -1 \end{bmatrix} \) normalizes to \( \mathbf{u_2} = \frac{1}{\sqrt{2}} \begin{bmatrix} 1 \ -1 \end{bmatrix} \).

Upon obtaining the orthogonal eigenvectors, these help in forming matrix \( P \), a pillar in the orthogonalization process.

The characteristic equation is foundational to matrix diagonalization. It is employed to determine eigenvalues of a matrix. The equation is derived by setting the determinant of \( A - \lambda I \) to zero, where \( \lambda \) represents the eigenvalue.

The given problem provides:

• Matrix \( A = \begin{bmatrix} a & b \ b & a \end{bmatrix} \), which is an example of a 2x2 symmetric matrix.
• Setting its determinant \( \det(A - \lambda I) = (a - \lambda)^2 - b^2 = 0 \) leads to solving \( (a - \lambda)^2 = b^2 \).
• This equation simplifies into the direct eigenvalues \( \lambda_1 = a + b \) and \( \lambda_2 = a - b \), crucial for the subsequent steps.

Understanding this step solidifies the basis for finding both eigenvalues and the resulting eigenvectors, opening the door to matrix transformations and other linear algebra applications.