Scalable Modeling and Synthesis of On-Chip Spiral Inductors

Abstract

In this paper, an efficient synthesis for integrated spiral inductors based on particle swarm optimization is presented. The synthesis/optimization process aims to determine the inductor layout parameters that maximize the quality factor while considering the imposed restrictions (project specifications). The spiral inductors are modeled using a double-π scalable model, in which the design parameters of inductor are taken as input. All the elements of the scalable model are calculated from the scalable expressions and process parameters. Particle swarm optimization is used to explore the design space and resolution of the optimization problem. Three design examples of inductors with inductance values of 3 nH, 5.5 nH and 8 nH, and with operating frequency of 2 GHz, 3.5 GHz and 5 GHz are used to evaluate the synthesis process. The obtained results confirm the validity of the adopted methodology and the PSO efficiency in determining the most adequate inductor structure to the project specifications.