Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

$$ A=\begin{bmatrix} a & b & b\\ b & a & b\\ b & b & a\\ \end{bmatrix} and \space B=\begin{bmatrix} 0 & A\\ A & 0\\ \end{bmatrix} $$

What are the eigen values and eigen vectors of B? Any help is appreciated.

share|cite|improve this question

For each eigenvalue $\lambda_n$ and eigenvector $x_n$ of a matrix $A$, the matrix $\pmatrix{0&A\\A&0}$ has eigenvalues $\pm\lambda_n$ with eigenvectors $\pmatrix{x_n\\x_n}$ and $\pmatrix{x_n\\-x_n}$. Your matrix $A$ becomes a constant matrix if we subtract $a-b$ on the diagonal, so it has eigenvalue $a-b$ with eigenspace the orthogonal complement of a constant vector (i.e. all vectors that sum to zero); since it's symmetric, the remaining eigenvector must be orthogonal to this eigenspace, so it's the constant vector, and the eigenvalue is $a+2b$ by inspection.

share|cite|improve this answer
Thanks. But my B matrix is a little different from what you assumed it to be. – saroj Aug 23 '12 at 5:14
@saroj: No, it's just a little different from how I accidentally wrote it :-) I've corrected that; the rest of the answer should be OK. – joriki Aug 23 '12 at 6:05

As an addition to the below answer, you may also want to notice that $A$ is a circulant matrix. That is, each row/column is a circularly shifted version of the row/column next to it. Hence, its eigenvectors are same as the $3$-point discrete Fourier transform (DFT) basis. The corresponding eigenvalues can be found out by taking the inner product of the eigenvector with the first row of $A$.

You can read more on circulants here.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.