Waveform relaxation based analysis of noise propagation in power distribution networks

In this paper, a waveform relaxation algorithm is presented to efficiently model power distribution networks. The proposed algorithm is based on physically partitioning the large circuit into smaller disjoint subcircuits. A key feature of the partitioning scheme is that it ensures that the noise injected by each transient source is localized within the subcircuit where the source resides thereby leading to efficient convergence of the algorithm. The iterative solution of the subcircuits is parallelizable and scales efficiently with the number of processors. A numerical example has been provided to illustrate the validity of the proposed algorithm over full SPICE simulations.