Course #06

Linearisation and Modelling Techniques for RF Power Amplifiers in Modern Communications Systems

October 4-7, 2010. Copenhagen, Denmark

INSTRUCTORS
Professor Steve C Cripps, University of Cardiff and Hywave Associates, Somerset, UK
Dr John Wood, Freescale Semiconductor, Inc., Tempe, AZ, USA

TECHNOLOGY FOCUS
The increasing use of linearisation techniques, and especially the emergence of high speed digital processing as an enabling technology to implement predistortion on the PA input signal, represent an important paradigm shift in PA design. The PA component can now be designed with more emphasis on efficiency, without the traditional constraints of meeting stringent linearity specs simultaneously. Maximising the utility of a lineariser in order to obtain optimum efficiency has thus become a new subject area in modern RF PA design 

COURSE CONTENT
This course takes a "system-level" approach to the linearization and behavioural modelling of RF Power Amplifiers. As such the course will not cover RF PA design in detail and can be considered as complementary to CEI-Europe course #08. Special emphasis is given to a detailed treatment of PA modelling, both as a means of allowing more meaningful system level simulations, and also as a necessary starting point to the development of advanced predistortion algorithms for PA linearization.

The course is dealing with system level issues in the specification and performance impact of RF power amplifiers in wireless, satcom, and microwave applications. It features in-depth treatment of behavioural modeling of PA non-linear effects. The main focus is to understand, characterize and model the non-linear behaviour of RFPAs, and to demonstrate the evolution of these models into successful predistortion algorithms. Particular emphasis will be given to the treatment of memory effects.

WHO SHOULD ATTEND
This course presents an overview, fundamentals, theory, practical and advanced power amplifier design, which will be of interest to RFPA designers and RF system designers. DSP developers who will be implementing the linearization algorithms will find this course beneficial, providing the necessary background knowledge in RFPA techniques and non-linear properties. 

BENEFITS
Enhance your understanding of:

• Power amplifier basic concepts, classes of operation, stability, linearity, bias technique
• PA Linearisation techniques
• PA non-linearities including memory effects
• Behavioural modeling of PAs, including memory effects
• Use of behavioral models to formulate accurate predistortion algorithms
• Application of pre-distortion linearization to RF PAs

Monday - Steve C. Cripps

Power Amplifier Overview 
Day 1 will start with a brief review of RFPA technology and basic RFPA theory, focussing on mode classifications widely used in PA literature. This will be a functional overview of the subject, circuit topologies will not be covered. The second part of Day 1 will focus on various system level approaches to PA efficiency enhancement.

• RF PA Semiconductor Technologies
• PA Modes A to Z
• Efficiency Enhancement Techniques (Doherty, EER, ET, Outphasing)

Tuesday - Steve C. Cripps / John Wood

Introduction to PA Non-linearity Effects - Steve C.Cripps 
Day 2 will start with an introduction to non-linear effects in RFPAs. The relationship between gain non-linearities and system level specifications such as ACP and EVM will be discussed. Memory effects will be discussed from a practical viewpoint. Later in Day 2 the modelling part of the course will start, defining various model types and their use in modern digital communications systems.

• AM-AM, AM-PM
• IM and ACP for Typical Digital Modulation Environments
• Memory Effects.

Nonlinear Modelling Basics 

• Compact Models: Applications and advantages
• Usefulness of Behavioural Models and Their Applications
• Review of Wireless Communications Signals and Systems, and Implications for Modelling.

Wednesday - John Wood

Advanced Behavioural Modeling 
Day 3 will continue the discussion on behavioural modeling, getting into some mathematical detail with the power series, the general Volterra formulation and its approach to model memory effects.

• System Identification Techniques: Mathematical models, including power series
• Volterra Series Formulations
• Artificial Neural Networks
• Behavioural Model Implementations
• Memory Effects.

Thursday - John Wood

Linearization
Day 4 will focus on the theory and implementation of digital predistortion, building on some of the models that have been introduced on previous days. "Linearizability" will be defined and several case studies will be considered.

• 'Linearizability' 
• Digital Pre-distortion: Implementation, Algorithms, Reprise of Memory Effects
• Case Studies in Behavioural Model Generation 
• RF PA Linearization Using DPD.

Regular Course Fee: 2490

Early Registration Course Fee: 2240
This applies to firm registrations received 2 months before course start. 

University Student and Faculty Rate:
Two university participants are welcome to attend for one course fee if payment is to be made from university funds.

Deliverables:
The course fee covers tuition, course material, and the day conference packages (morning/afternoon refreshments, and lunches) paid on your behalf to the course venue. Accommodation is not included.