Enhancing RF Energy Harvesting and Wireless Power Transfer with GAN-Optimized 3D Quasi-Yagi Antenna

Radio Frequency Energy Harvesting (RFEH) and Wireless Power Transfer (WPT) are increasingly popular for powering low-power devices, with the design of receiving antennas being a key factor. This paper presents a cube-shaped 3D antenna, innovatively derived from the quasi-Yagi antenna (QYA) design and optimized for operation at 915 MHz. Unlike traditional 2D and directional antennas, this 3D design offers multidirectional reception, which is crucial for efficiently harvesting varying levels of dedicated and ambient energy. Using CST Studio tools for simulation, the antenna, based on an optimized 2D QYA model (Type D with 6.24 dBi gain and 43.87 dB return loss), achieved a high gain of 11.43 dBi, a return loss of 45.61 dB, and a bandwidth of 25 MHz, as verified by the N9912A FieldFox HandHeld RF Analyzer and ME1310 Antenna Training Kit (3D) Dream Catcher. Generative Adversarial Networks (GANs) were utilized for data augmentation and optimization, reducing the reliance on extensive data collection. Additionally, the integration of this 3D antenna with an impedance-matching network and subsequent experimental harvesting with a 1.5 kΩ resistor demonstrate its practicality. We thus propose a versatile and efficient RFEH solution suitable for a range of RF sources and applications.