Robust CMOS CNN implementation with respect to manufacturing inaccuracies
A. Paasio, Adam Dawidziuk, Kari A. I. Halonen, Veikko Porra · 2002
A new idea of cellular neural network VLSI implementation based on a high gain sigmoid function is shown. It has been proved in this paper that this method is robust to the parameter inaccuracies. Conclusions lead to some suggestions and restrictions for template tolerances. An example of VLSI implementation-the test chip consisting of 16/spl times/16 cells in 1.2 /spl mu/m CMOS with 19 fully programmable coefficients has been fabricated. Measurements indicate good efficiency of this implementation.