AUTHORS: A.K. Singh, Azam Beg and P.W.C. Prasad
PUBLICATION/VENUE: International Conference for Engineering and ICT, Nov 2007, pp. 101-105.
The main idea of this paper is to use the neural network (NN) modeling methodology to learn the behavior of a path length property (longest path length/LPL) of binary decision diagrams (BDDs). The proposed method utilizes data derived from Monte-Carlo simulations for different number of variables and product terms (min-terms). The NN modelís (NNMís) predictive ability is proven by comparing the actual values of LPL and NNM predictions, for ISCAS benchmark circuits. The model allows design feasibility and performance to be analyzed prior to the circuit implementation.
CITATIONS: Google | Yahoo