Inverse Design of Compact and Wideband Inverted Doherty Power Amplifiers Using Deep Learning
arXiv:2606.27002v1 Announce Type: cross Abstract: This paper presents a deep learning-assisted methodology for the inverse synthesis of a compact, wideband inverted Doherty power amplifier (PA). Convolutional neural networks (CNNs) and genetic algorithms (GAs) are jointly employed to generate pixelated Doherty combiner networks that integrate load modulation, impedance matching, power combining, and phase compensation into a single structure. As a proof of concept, we design and fabricate a GaN HEMT Doherty PA with a pixelated output combiner. The prototype achieves a measured peak drain effic
The increasing demands for efficiency and performance in communication systems, coupled with advancements in AI and deep learning, makes this a timely development.
This development offers a method to significantly improve the efficiency, compactness, and bandwidth of power amplifiers, crucial components in various communication technologies.
The conventional design process for high-performance power amplifiers can be significantly accelerated and optimized through AI-driven methodologies, potentially leading to faster innovation cycles and better device characteristics.
- · Telecommunication companies
- · Semiconductor manufacturers
- · AI/ML design tool developers
- · Wireless communication sector
- · Traditional PA design methodologies
- · Companies without AI integration
Increased efficiency and reduced size in power amplifier components for radio frequency systems.
Faster development and deployment of next-generation wireless communication standards (e.g., beyond 5G/6G) due to improved hardware.
Lower energy consumption in data centres and mobile devices linked to more efficient power amplification, subtly impacting the energy bottleneck narrative.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI