Puneet Varma (Editor)

Spectral abscissa

Updated on
Edit
Like
Comment
Share on FacebookTweet on TwitterShare on LinkedInShare on Reddit

In mathematics, the spectral abscissa of a matrix or a bounded linear operator is the supremum among the real part of the elements in its spectrum, sometimes denoted as η ( A )

Matrices

Let λ1, ..., λs be the (real or complex) eigenvalues of a matrix ACn × n. Then its spectral abscissa is defined as:

η ( A ) = max i { R e ( λ i ) }

For example if the set of eigenvalues were = {1+3i,2+3i,4-2i}, then the Spectral abscissa in this case would be 4.

It is often used as a measure of stability in control theory, where a continuous system is stable if all its eigenvalues are located in the left half plane, i.e. η ( A ) < 0

References

Spectral abscissa Wikipedia


Similar Topics