Design of multioctave high-efficiency power amplifiers using stochastic reduced order models

Zhuang, Yuan and Fei, Zhouxiang and Chen, Anqi and Huang, Yi and Rabbi, Khondker and Zhou, Jiafeng (2018) Design of multioctave high-efficiency power amplifiers using stochastic reduced order models. IEEE Transactions on Microwave Theory and Techniques, 66 (2). ISSN 0018-9480

Text Zhuang_etal_ITMTT2017_Design_multioctave_high_efficiency_power_amplifiers_stochastic_reduced_order_models.pdf - Accepted Version Download (3MB)

Abstract

This paper presents a novel general design method of frequency varying impedance matching. The method is applied to design of a broadband high-efficiency power amplifier (PA). The proposed method defines the optimal impedance regions of a PA at several frequency sections over the operational frequency band. These regions contain the impedances that can achieve a high output power and a high-power added efficiency (PAE) simultaneously. A low-pass LC-ladder circuit is selected as the matching network (MN). The element values of the MN can be obtained using a synthesizing method based on stochastic reduced order models and Voronoi partition. The MN provides desired impedance in the predefined optimal impedance region at each frequency section. Thus, optimal output power and PAE of the PA can be achieved. To validate the proposed method, two eighth-order low-pass LC-ladder networks are designed as the input and output MNs, respectively. A gallium nitride (GaN) HEMT from Cree is employed as the active device. Packaging parasitic of the transistor has been taken into account. A PA is designed, fabricated, and measured. The measurement results show that the PA can achieve P1 dB PAE of better than 60% over a fractional bandwidth of 160% (0.2-1.8 GHz). The output power is 42-45 dBm (16-32 W), and the gain is 12-15 dB. The performance of the PA outperforms existing broadband highefficiency PAs in many aspects, which demonstrates the excellence of the proposed method.

Item Type:	Article
Faculty / Department:	Faculty of Human and Digital Sciences > School of Computer Science and the Environment
Depositing User:	Zhouxiang Fei
Date Deposited:	13 Feb 2025 12:23
Last Modified:	13 Feb 2025 12:23
URI:	https://hira.hope.ac.uk/id/eprint/4607

Actions (login required)

View Item