A Neural Network-Based Design Automation of a Second Generation Current Conveyor

Abstract

An artificial neural network approach for the automated design of a positive type second generation current conveyor is presented in this paper. A multi-layer perceptron structure is successfully employed to estimate the corresponding transistor dimensions for a given set of desired performance criteria of the circuit. Data generated by a circuit simulation program (SPICE) is used to train the artificial neural network. The excellent agreement between the desired specifications and the actual results from SPICE simulation results approves that neural networks are powerful tools for automated analog circuit sizing.