In almost all analog modeling algorithms, we solve a (non-)linear system they require at some point to solve , with given and . Depending on the size of the matrix and its characteristics, computing an inverse can be costly and may incur numerical problems. Let’s tackle cost in this discussion.
