Automatic Design of Power Amplifier Based on Machine-Learning-Assisted Optimization

This paper proposes an automated design process for power amplifiers based on simplified real frequency technique (SRFT)-driven machine learning assisted optimization (MLAO). The SRFT is introduced for initial design allocation, followed by the MLAO to automatically yield optimal design solution. To validate the proposed strategy, a power amplifier consisting of a 10 W GaN HEMT is designed for 1–2 GHz band frequency. The designed P A achieves a power added efficiency more than 65% with the output power from 39.4 dBm to 42.0 dBm across the band.