Analog synthesis of nonlinear functions based on fuzzy logic

摘要

A circuit for the analog implementation of nonlinear functions hasnbeen designed. It is based on the fuzzy-logic paradigm mapped into anmodular architecture. The basic building blocks rely on the large-signalncharacteristic of MOS differential pairs to design bell-shaped basisnfunctions and combine them with the center-of-gravity method. The actualninput/output (I/O) characteristics are defined by applying propernvoltage levels to the high-impedance input nodes of each module. Both anfield-programmable processor, where the programming configuration isnstored in a digital RAM, and a tool for generating the layout ofndedicated circuits, where this connection is hardwired, have beenndesigned using a CMOS 0.7 Μm n-well technology. A software layerncomputes the programming values which allow the circuit response tonapproximate the target I/O relationship. Experiments show that thenapproximation accuracy is in the range of a few percent r.m.s. of thencircuit output range. The application of the system to the synthesis ofna nonlinear control law for a DC-DC power converter is discussed