Solving functional postfabrication tuning of circuits with symmetry problems by introducing an additional feature to neural network learning

Abstract

The authors define a new parameter, the combined improvement figure (IM), which is used with backpropagation neural network learning, and provides better discrimination in postfabrication tuning of circuits with symmetry property problems. Results from examples show that there is a substantial improvement in identifying correct tuning parameter and tuning levels when this figure is included.