Title :
The application of neural networks to high speed interconnect simulation
Author :
Veluswami, Anand ; Zaabab, A. Hafid ; Zhang, Qi-Jun ; Nakhla, Michel S.
Author_Institution :
Dept. of Electron., Carleton Univ., Ottawa, Ont., Canada
Abstract :
A new CAD tool for high speed VLSI interconnect simulation is presented. Neural networks are capable of modeling the relationship between an interconnect network and its associated signal integrity parameters in an efficient, yet simple manner, producing fast models ideally suited for CAD and optimization routines. The resultant models have high accuracy and low run times, and give substantial speedup over existing simulation techniques
Keywords :
VLSI; circuit CAD; circuit analysis computing; circuit optimisation; feedforward neural nets; integrated circuit design; integrated circuit interconnections; integrated circuit modelling; CAD tool; high speed VLSI interconnect simulation; high speed interconnect simulation; interconnect network; neural networks; optimization routines; run times; signal integrity parameter; simulation techniques; speedup; Circuit simulation; Computational modeling; Crosstalk; Design automation; Design optimization; Feedforward neural networks; High performance computing; Integrated circuit interconnections; Network topology; Neural networks;
Conference_Titel :
Electrical and Computer Engineering, 1995. Canadian Conference on
Conference_Location :
Montreal, Que.
Print_ISBN :
0-7803-2766-7
DOI :
10.1109/CCECE.1995.528155
