Title :
Artificial neural networks for reverse engineering bipolar transistors
Author :
Ferguson, Ryan ; Roulston, David J.
Author_Institution :
Dept. of Electr. & Comput. Eng., Waterloo Univ., Ont., Canada
Abstract :
In this paper we report on progress made in developing an artificial neural network which can reverse engineer the physical descriptions of bipolar transistors from a complete set of electrical data. The neural network tool REED (Rapid Engineering of Electron Devices) is used to perform a series of SPICE to BIPOLE3 mappings
Keywords :
SPICE; bipolar transistors; learning (artificial intelligence); neural nets; reverse engineering; semiconductor device models; semiconductor process modelling; BIPOLE3; REED; SPICE; artificial neural network; bipolar transistors; electrical data; impurity profile approximation; neural network tool; rapid engineering of electron devices; reverse engineering; Artificial neural networks; Backpropagation; Bipolar transistors; Data engineering; Electrons; Feedforward neural networks; Feedforward systems; Neural networks; Reverse engineering; SPICE;
Conference_Titel :
Microelectronics, 1997. Proceedings., 1997 21st International Conference on
Conference_Location :
Nis
Print_ISBN :
0-7803-3664-X
DOI :
10.1109/ICMEL.1997.632868
