Skip to main content
Open Access Publications from the University of California

UC San Diego

UC San Diego Electronic Theses and Dissertations bannerUC San Diego

A Highly Efficient And Linear RF Power Amplifier For Mobile Terminal Applications /


The radio frequency (RF) power amplifier (PA) is the last block in a transmitter chain. It amplifies the signal to the target power and drives the antenna. The power amplifier consumes the largest portion of the transmitter current consumption budget, and any power saving in this block will significantly improve the overall system efficiency. This is especially critical in battery- operated portable wireless communication systems such as cellular phones, PDA's and laptops. There is a trade-off between efficiency and linearity in power amplifiers. Non- switching power amplifiers efficiency increases as the output power increases, but so does the amplifier distortion. This is more problematic in modern wireless communication systems, where spectrally efficient and high data rate modulations are used and the linearity requirement is hard to meet. The two common approaches of backing-off the output power and adding a linearization scheme have their own challenges. The first one results in efficiency loss and the second one adds to the system complexity and has design challenges for wideband applications. The design of highly efficient and linear RF power amplifiers has been the subject of several studies. Different techniques have been proposed to overcome the challenge. Dynamic control of the power amplifier quiescent current, dynamic control of the load impedance, output harmonic control and dynamic supply voltage control (envelope tracking) are the popular proposed techniques. Despite the fact that the envelope tracking technique has gained momentum as an attractive efficiency enhancement method for handset applications, its implementation still faces challenges. This technique needs a high efficiency envelope amplifier to achieve good overall efficiency. The design of a small, efficient and wideband envelope amplifier is very challenging. Usually these amplifiers require external components and if they are in the switching mode they can add disturbance to the rest of the system. This research focuses on a technique that overcomes this main challenge of the ET amplifier design and is organized as follows : Chapter 1 is the introduction and discusses the motivation of this research and some of the prior art. Chapter 2 explains the proposed technique to enhance the RF power amplifiers efficiency in high peak-to-average power ratio applications. This technique is based on controlling the baseband drain impedance by adding an envelope termination to the PA supply, and applying the baseband envelope signal to the input. As a result, the amplifier operates closer to its saturated region for all envelope amplitudes, and its efficiency is improved. A digital predistortion scheme is implemented to compensate for the linearity degradation of the proposed technique. A 1.95GHz HFET power amplifier exhibits an improvement in peak PAE from 40% to 56% for a two-tone input, from 33% to 42% for an uplink WCDMA with one dedicated physical data channel and from 27% to 32% for an uplink WCDMA with six dedicated physical data channels using the proposed technique. Chapter 3 shows some improvements to the proposed technique including adding envelope equalization. The envelope equalization also improves the amplifier linearity since it reduces the distortion from clipping the output signal due to non- ideal dynamic supply. The linearity of the implemented amplifier is studied in a large-signal fashion. This technique improves the maximum efficiency from 28% to 40% for an uplink WCDMA signal with six dedicated physical data channels and the maximum linear efficiency from 21% to 28%. Also, a scheme to vary the DC power supply with the average power to maintain high efficiency down to a low average power region is proposed. Chapter 4 studies envelope feedback systems stability. The stability criteria are derived based on the Lyapunov stability theorem for time-varying systems and the tool of linear matrix inequality (LMI). The effects of system parameters on the stability are investigated. The system is simulated in Simulink and implemented, and stability boundaries predicted by LMI are in good agreement with the simulation and measurement results. Chapter 5 concludes the dissertation and suggests some future work

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View