A Nonlinear Dynamic Model for Performance Analysis of Large-Signal Amplifiers in Communication Systems

Mirri, D.; Filicori, F.; Iuculano, G.; Pasini, G.

doi:10.1109/tim.2003.822714

Cited by 33 publications

(23 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is similar to the Modified Volterra Series [8], but retains the property of linearity in the parameters of the model, as for the classical Volterra series. Based on this new representation, an effective model order reduction method was proposed, called Dynamic Deviation Reduction in which higher order dynamics are removed since the effects of nonlinear dynamics tend to fade with increasing order in many real power amplifiers.…”

Section: B Dynamic Deviation Reduction-based Volterra Model (Ucd)mentioning

confidence: 90%

System level modelling of RF power amplifiers based on large-signal measurements

Fager

Lavrador²,

Myslinski³

et al. 2007

2007 European Microwave Integrated Circuit Conference

View full text Add to dashboard Cite

This paper demonstrates the extraction of three different system level behavioural models of power amplifiers from large signal time domain measurements. The modelling results are evaluated using a common independent CDMA2000SR1 telecommunication input signal. The capability of the models to predict long term memory effects is also evaluated by artificially modifying the bias network in one of the PA prototypes. I. INTRODUCTIONThe integration of more functionality, and thus complexity, in wireless transceivers, turned the traditional circuit level based simulation highly inefficient, which asked for system level analysis, and so reduced order models. Unfortunately, the simplicity of such reduced order models is only apparent when we have interacting nonlinearity and dynamics. This is the case of microwave power amplifiers (PAs) where an increasing interest has been seen to look for compact models that can reproduce their observable input-output behaviour without the need for the complete internal description [1]. Indeed, this model simplicity has a high cost in predictive capability and the measurements required for parameter set extraction, since we are no longer relying on any a-priori knowledge of the PA physics.Therefore, a great care is paid to the selection of the model format, since not all mathematical formulations are able to reproduce every input-output mapping encountered in nature. For example, no one would try to use a linear S-parameter matrix to reproduce nonlinear behaviour, nor would one use a polynomial, or any other static function, to account for the PA fast (time constants of the order of the carrier period) or slow dynamics (time constants of the order of the inverse of the modulated signal bandwidth). Nonlinear system identification theories have shown that the model of Fig. 1 is one of these possible general formulations as long as (i) enough delays are available in the first linear dynamic block and (ii) the nonlinear static block is a universal approximator [1].Also, because the model is built from (behavioural) inputoutput observations, it should be intuitive that its accuracy degrades when its response is tested for signal excitations that are farther and farther away from the signal, or signal class, used to extract it. For example, we should not expect that such a model can predict the response to a signal that exposes some PA property never tested during model extraction. Hence, excitation selection for model extraction and validation plays a vital role on the final model quality [2].

show abstract

Section: B Dynamic Deviation Reduction-based Volterra Model (Ucd)mentioning

confidence: 90%

System level modelling of RF power amplifiers based on large-signal measurements

Fager

Lavrador²,

Myslinski³

et al. 2007

2007 European Microwave Integrated Circuit Conference

View full text Add to dashboard Cite

show abstract

“…They are also classified into two categories: neural-network based models which consist of artificial neuralnetworks [16] and dynamic fuzzy neural networks [17], and Volterra series [7,8] based models such as memory polynomial (MP) [9], Volterra with dynamic deviation reduction [18], generalized memory polynomial [19].…”

Section: Power Amplifier Behavioral Modelingmentioning

confidence: 99%

Mixed-Signal Multimode Radio Software/Hardware Development Platform

et al. 2013

View full text Add to dashboard Cite

Radio frequency power amplifiers (PAs) are the most challenging part of the design of radio systems since they dictate the overall system's performance in terms of power efficiency and distortion generation. The performance is further challenged by modern modulation schemes which are characterized by highly varying signal envelopes. In order to meet the spectrum mask requirements, PAs are usually operated at high power back-off to ensure linearity, at the cost of efficiency. To tackle this issue, many efficiency enhancement techniques have been presented in the literature. In fact, these techniques do increase the PA power efficiency at back-off, however, efficiency enhancement techniques do not ensure the linearity of the PA. Furthermore, these techniques may lead to additional distortion. On the other hand, several linearization techniques have been developed to mitigate the PA nonlinearity problem and allow the PA to operate at less back-off. Digital Pre-Distortion (DPD) technique is gaining more attention, as compared to other linearization techniques, thanks to its simple concept and advancements in digital signal processors (DSP) and signal converters. DPD technique consists of introducing a nonlinear function before the PA so that the overall cascaded system behaves linearly. It was clear from the literature that this technique showed good performance. Yet, it has primarily been validated using commercial test equipment, which has good capabilities, and far from the real world environment in which this technique would be implemented. Indeed, DPDs would need to be implemented in signal processors characterised by limited resources and computational accuracy. This thesis presents an implementation of several DPD models, namely look-up table (LUT), memoryless polynomial and memory polynomial (MP), on a field programmable gate array (FPGA). A novel model reformulation made this implementation possible in fixed-point arithmetic. Measurements were collected to validate the DPD models' implementation and an improvement of the signal quality was recorded in terms of error vector magnitude (EVM) and adjacent channel leakage ratio (ACLR).As many wireless access technologies must continue to coexist, multi-standard radio systems are required to reduce the cost while maintaining the interoperability. This thesis presents a development platform for multimode radio which comprises mixed-signal modules. The platform provides the capacity for hardware and software development. In fact, the FPGA under investigation allowed for the implementation of a baseband transceiver and DPD schemes. In addition, a software tool was developed as a dashboard to control and monitor the system. The radio system in the platform was optimized through the equalization of the feedback receiver frequency response performed through a simultaneous measurement of the amplitude ripple of the transmitter and receiver. Furthermore, a phase-coherent frequency synthesizer was designed to bring more flexibility by allowing the transmitter's ca...

show abstract

“…Modified/Dynamic Volterra series The Modified Volterra Series [12][13][14], or Dynamic Volterra Series [15], is based on introducing the dynamic deviation function e(n,i): e(n, i) = x(n -i) -x(n) (8) which represents the deviation of the delayed input signal x(n-i) with respect to the current input x(n). Substituting (8) in (1), the input/output relationship for a nonlinear system with memory can be described as (9) where yS(n) is the static part and can be expressed as a power series of the current input signal x(n): This model can be implemented by a block which is similar to a Finite Impulse Response (FIR) filter, but odd-order polynomials are used instead of the linear gain taps of the filter.…”

Section: B Direct Pruningmentioning

confidence: 99%

“…The most important property of this modified series is that it separates the purely static effects from the dynamic ones, which are intimately mixed together in the classical series. However, this modified Volterra series loses the property of linearity with respect to model parameters, which means that the output of the model is no longer linear with respect to the coefficients [14]. This leads to the consequence that models of this kind cannot be extracted in a direct way using established linear system estimation procedures y(n) = y, (n) + y, (n) such as the least squares techniques, as is usual in the classical case.…”

Section: B Direct Pruningmentioning

confidence: 99%

An Overview of Volterra Series Based Behavioral Modeling of RF/Microwave Power Amplifiers

Zhu

Brazil

2006

2006 IEEE Annual Wireless and Microwave Technology Conference

View full text Add to dashboard Cite

show abstract

A Nonlinear Dynamic Model for Performance Analysis of Large-Signal Amplifiers in Communication Systems

Cited by 33 publications

References 6 publications

System level modelling of RF power amplifiers based on large-signal measurements

System level modelling of RF power amplifiers based on large-signal measurements

Mixed-Signal Multimode Radio Software/Hardware Development Platform

An Overview of Volterra Series Based Behavioral Modeling of RF/Microwave Power Amplifiers

Contact Info

Product

Resources

About