An analytical approach based on neural computation to estimate the lifetime of deep submicron MOSFETs

The VLSI-component industry requires more financial investment than ever in order to measure the growing sophistication of the manufactured products and for the equipment necessary to their development. So, the modelling of electric components constitutes a research field that is currently very important throughout the world. To continue this evolution, the existing models must be improved and new models have to be developed. Hence, we regularly see improvements of simulation software. In this paper, we present the applicability of artificial neural networks for the development of an analytical approach allowing the evaluation of the time degradation at deep submicron level of MOSFETs devices. This approach can be implemented in electronics simulators (SPICE, PSPICE, CADENCE, ...). Our results are compared with the experimental ones, analysed and discussed in order to draw some useful information and decisive conclusions about the VLSI technology.

[1]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[2]  H. Lifka,et al.  Three hot-carrier degradation mechanisms in deep-submicron PMOSFET's , 1995 .

[3]  E. Vittoz,et al.  An analytical MOS transistor model valid in all regions of operation and dedicated to low-voltage and low-current applications , 1995 .

[4]  E. Takeda,et al.  An empirical model for device degradation due to hot-carrier injection , 1983, IEEE Electron Device Letters.

[5]  Kang-Deog Suh,et al.  Impact of floating gate dry etching on erase characteristics in NOR flash memory , 2002, IEEE Electron Device Letters.

[6]  B. Yegnanarayana,et al.  Artificial Neural Networks , 2004 .

[7]  J. Pelloie,et al.  Hot-carrier effects and reliable lifetime prediction in deep submicron N- and P-channel SOI MOSFETs , 1998 .

[8]  A. Bravaix,et al.  Lifetime prediction methods for p-MOSFET's: a comparative study of standard and charge-pumping lifetime criteria , 1995 .

[9]  I. Dimopoulos,et al.  Application of neural networks to modelling nonlinear relationships in ecology , 1996 .

[10]  T. Manku,et al.  Microwave CMOS-device physics and design , 1999, IEEE J. Solid State Circuits.

[11]  Chenming Hu,et al.  Hot-electron-induced MOSFET degradation—Model, monitor, and improvement , 1985, IEEE Transactions on Electron Devices.

[12]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.