Single-objective front optimization
Eduardo J. Solteiro Pires, Luís P. N. Mendes, Paulo Moura Oliveira, J. A. Tenreiro Machado, N. M. Fonseca Ferreira, Joao Caldinhas Vaz, Maria Joao Rosario · 2008
Obtaining a well distributed non-dominated Pareto front is one of the key issues in multi-objective optimization algorithms. This paper proposes an algorithm which promotes well distributed non-dominated fronts in the parameters space when a single-objective function is optimized. The proposed technique is described and tested in an automated synthesis circuit design problem. The project consists in designing CMOS radio-frequency and microwave binaryweighted differential switched capacitor arrays (RFDSCAs) from user top-level specification to component size. The genetic synthesis tool optimizes a fitness function based on the performance parameter of the RFDSCAs. To validate the proposed design methodology, a CMOS RFDSCA is synthesized, using a 0.25 μm BiCMOS technology, and verified by the SpectreRF simulator of the Cadence design environment. The results show that the synthesis and simulation outcomes are in very good agreement.