An elitist distributed particle swarm algorithm for RFIC optimization

An RF IC optimization methodology based on an elitist distributed particle swarm optimization algorithm is presented. By including a Pareto ranking mechanism and elitism in the algorithm, design alternatives and tradeoff information are provided with high efficiency. Post-optimization Monte-Carlo simulations are performed to assess first-order yield performance and aid in the selection of the final design. The approach is validated through the synthesis of a 5.2GHz direct-conversion front-end in 180nm CMOS.