In this paper, artificial neural networks are applied to the modeling of the frequency-dependent parameters of lossy interconnects. Neural networks are an efficient mathematical tool capable of mapping the relationship between the physical parameters of interconnects and their per-unit-length parameters over a large range of frequency. They offer high on-line speed and are ideally suited for use in iterative computer-aided design and optimization techniques.

Additional Metadata
Conference Proceedings of the 1997 IEEE Multi-Chip Module Conference
Citation
Veluswami, Anand, Nakhla, M.S, & Zhang, Q.J. (1997). Modeling the frequency-dependent parameters of high-speed interconnects: a neural network approach. Proceedings of the IEEE Multi-Chip Module Conference.