Design automation for load modulated GaN power amplifiers

The shift to 6G demands the use and integration of advanced technology such as GaN as it stands out for its high efficiency, power density, and performance in high-frequency applications. This is true when designing high-frequency high-efficiency power amplifiers (PA).

State-of-the-art PAs often use load modulation techniques to increase the power efficiency of the amplifier. Combining the load modulation techniques - such as Load Modulated Balanced Amplifiers (LMBA), Outphasing PA (OPA), and Doherty PA (DPA) - with the black box models extracted using load-pull simulations/measurements is a general and effective strategy to design the PA at hand. However, the synthesis of the passive distribution and combining networks obtained from this black-box approach is currently still a manual engineering step.

The aim of this PhD is to

• unify the black-box load-modulated PA design approaches to cover a wide range of existing topologies (LMBA, OPA, DPA),
• use load-pull simulations and measurements to determine the impedance matrixes for both distribution and combining networks,
• research the automatic synthesis of distribution and combining networks by combining filter synthesis and AI techniques such as Reinforcement Learning, and
• apply the developed techniques on designs in advanced GaN technologies.

Hence, this PhD must result in a general black-box design methodology that is demonstrated on simulated and measured GaN PA designs (LMBA, OPA, DPA).