Conference Paper

Time-domain neural network characterization for dynamic behavioral models of power amplifiers

Dpt. Ingegneria Elettronica, Univ. Tor Vergata, Rome, Italy
Conference: Gallium Arsenide and Other Semiconductor Application Symposium, 2005. EGAAS 2005. European
Source: IEEE Xplore

ABSTRACT This paper presents a black-box model that can be applied to characterize the nonlinear dynamic behavior of power amplifiers. We show that time-delay feed-forward neural networks can be used to make a large-signal input-output time-domain characterization, and to provide an analytical form to predict the amplifier response to multitone excitations. Furthermore, a new technique to immediately extract Volterra series models from the neural network parameters has been described. An experiment based on a power amplifier, characterized with a two-tone power swept stimulus to extract the behavioral model, validated with spectra measurements, is demonstrated.

0 Bookmarks
 · 
83 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: Digital predistortion is used as a compensation technique in communication systems to minimize the effect of power amplifier non-linearity while increasing its operational efficiency. Prior linearization schemes use the receiver chain to feed data to the baseband processor that adaptively adjusts the predistortion coefficients. The procedure is iterative, requires many test applications, is sensitive to receiver quality, and is not suitable for RF communication front ends in which the mixer, LNA and PA are dynamically reconfigured to adapt to changing operating conditions. In the proposed scheme, a single multi-sine diagnostic test is applied from the baseband, and the response of the transmitter is captured via a response envelope detector. A novel unified methodology for co-tuning predistortion coefficients along with the PA bias voltage based on response diagnosis is proposed, thereby, enabling the transmitter to operate at high efficiency and linearity.
    25th IEEE VLSI Test Symposium (VTS 2007), 6-10 May 2007, Berkeley, California, USA; 01/2007
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a new approach to build RF dynamic behavioral models, based on time-delay neural networks (TDNNs), suitable for FET devices, and capable to identify the working class and to characterize both short- and long-term device memory, through a time-domain training procedure, for a wide range of input power levels. The presented model has been effectively applied to GaN-based devices, working in class A, AB and B
    European Microwave Integrated Circuits Conference, 2006. The 1st; 10/2006
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: System-level models simplify the analysis of complex RF systems, such as transmission-reception modules, by expressing global input-output relationships. However, the development of high RF power models for nonlinear subsystems requires the prediction of the distortion induced by low-frequency memory effects such as self-heating effects. In this framework, we present a new electrothermal behavioral model for power amplifiers. This global model is based on the coupling between a behavioral electrical model derived from the transistor-level description of the amplifier and a thermal reduced model. This model, implemented into a circuit simulator, allows to predict the impact of the thermal effects in pulsed RF mode thanks to an envelope transient analysis. This approach has also been validated by measurements.
    IEEE Transactions on Microwave Theory and Techniques 12/2007; · 2.23 Impact Factor

Full-text (2 Sources)

View
31 Downloads
Available from
May 27, 2014