Analog Circuit Design Automation Using Neural Network-Based Two-Level Genetic Programming

The design of analog circuits starts with a high-level statement of the circuit's desired behavior and requires creating a circuit that satisfies the specified design goals. The difficulty of the problem of analog circuit design is well known, and there is no previously known general automated technique to design an analog circuit from a high-level statement of the circuit's desired behavior. This paper proposes a two-layer evolutionary scheme based on genetic programming (GP) and neural network (NN), which uses a divide-and-conquer approach to design the analog circuits. Corresponding to the NN-TLGP, a new representation of circuit has been proposed and it is more helpful to generate expectant circuit graphs. This algorithm can perform the circuits with dynamical size, circuit topology, and component values. The experimental results on the two design work show that this algorithm is efficient