ControlRank
The
ControlRank
index is a spectral measure designed to quantify node importance in both static and dynamic networks under control settings [2]. ControlRank is based on linear control theory and evaluates the influence of a node by analyzing the minimum eigenvalues of matrices derived from the network's Laplacian. A larger minimum eigenvalue indicates that the node contributes more to the network’s overall stability and accelerates convergence dynamics.
The ControlRank index of node \(i\) is defined as
\[
c_{ControlRank}(i) = λ_i(L^s),
\]
where \(λ_i(L^s)\) is the minimum eigenvalue of the principal minor of \((L + L^T)/2\) obtained by removing the \(i\)-th row and column from the Laplacian matrix \(L\).