Analog circuit synthesis using two-port theory and genetic programming

Abstract

This paper describes a novel approach for analog electronic circuits synthesis using two-port theory and genetic programming. In contrast to other approaches, circuits are modeled as two-port networks which enables a straight forward encoding and evaluation of the circuit's structure. In this way, a number of active and passive linear circuits were synthesized and optimised for multiple objectives. The algorithm output comprises both evolved topologies and the corresponding component values.