Optimization of RF/microwave multichip module performance based on neural models of passives and interconnects

Widely used in multichip modules, distributed passive devices are subject to intensive research for efficient modeling in the RF/microwave frequency range. This paper introduces the concept of generation of accurate and fast neural models for passives that efficiently integrate high-frequency electromagnetic effects. By optimizing the circuit performance, the proposed approach can automatically predict not only the optimum geometrical parameters of passives but also their optimum location in the chip layout. Applications using neural models implemented in commercial circuit simulators are presented.