For mmWave Integrated Circuit (IC) design, co-integration of passives can reduce size and power consumption, increase reliability, and reduce overall cost. However, skin and proximity effects in the metallization are aggravated at mmWave frequencies, resulting in increased attenuation and degradation of overall performance. Furthermore, tight integration of passive components (to reduce the die size) poses additional design challenges due to the complicated electromagnetic couplings between the components in close proximity. Other parasitic effects such as dummy metal fill parasitics and substrate eddy current loss, the impact of process variability and uncertainty are also more prominent at mmWave frequencies.
In this thesis, scalable compact modeling techniques based on the Principle of Electromagnetic Similitude and additionally developed methods of complexity reduction are presented. Scalable and compact equivalent circuit models for on-chip microstrip and Coplanar Waveguide (CPW) are developed and validated by both electro-magnetic (EM) simulations and on-wafer measurements. Furthermore, a more general field-based scalable modeling approach for multi-conductor interconnects (e.g., coupled CPWs) and more complicated passives is presented. The field-based approach has been applied and validated for modeling mmWave inductors, including the impact of metal fills and substrate eddy-current effects. Scalable models to capture the magnetic coupling between spiral inductors are presented and a compact layout strategy for reduction of magnetic coupling is proposed. The scalable and compact models include multi-physics effects such as temperature dependency and can also be utilized to study the impact of the process variations efficiently.