Deep Learning-Driven Inverse Design of Doherty Power Amplifiers Using Pixelated Combiners and Dual-State Impedance Synthesis
The paper discusses using deep learning for the inverse design of Doherty Power Amplifiers, a complex circuit that requires managing load modulation and impedance matching. The team developed a methodology combining deep convolutional neural networks with a genetic algorithm. They successfully created GaN HEMT prototypes achieving saturated output power exceeding 44.2 dBm and peak drain efficiency over 71%.