In this paper we show that innovative physics-based simulations can be used for a comprehensive analysis of RF stages subject to random variations of technological parameters, including the computation of the average (deterministic) RF performance along with their statistical deviation. The variability analysis is addressed by means of the recently developed physicsbased sensitivity analysis of AC parameters through Green's functions [1], [2]. To demonstrate the technique, we address the analysis of a FinFET mixer exploiting an innovative Independent Gates topology, showing that a careful design allows to maximize the mixer conversion gain while minimizing its variability vs. several physical parameters, such as the gate length, oxide thickness and fin width.