| Review: |
Mathematical models are used to simulate or control dynamical systems whose future depends on their past evolution, such as the weather and VSLI circuits. This book deals with model reduction. The central problem is: Given a linear time-invariant input–output system defined by the convolution integral, a transfer function, a state–space representation, or their discrete-time counterparts, how can one approximate this system by a simpler system? Here the author presents approximation methods related to the singular value decomposition (SVD), to Krylov or moment matching methods, and to combinations of the two. He addresses the issue of model reduction and the resulting trade-offs between accuracy and complexity. |