The goals of RF power amplifier design are high efficiency and linearity. With modern cellular communications modulation formats such as LTE and WCDMA these goals are difficult to achieve simultaneously with traditional RF PAs, and high-efficiency architectures such as Doherty, Envelope Tracking, and so forth are becoming more commonplace.

These PAs require an additional linearization system to achieve the mandated linearity specifications.

The emergence of high-speed digital signal processing techniques has enabled the linearization to be accomplished in the digital signal domain, and digital pre-distortion (DPD) is now the preferred linearization technique.

Behavioral Modeling and Digital Predistortion of RF Power Amplifiers
Behavioral Modeling and Digital Predistortion of RF Power Amplifiers


This course explains the nonlinear behavior of RF power amplifiers, developing general modeling techniques to describe the nonlinearities and memory effects. A system-level approach to the modeling and linearization of the PA is adopted, and techniques for implementation of DPD in practical situations are described.

Upon completing the course, the participant will be able to:

  • Understand and describe the nonlinear behavior and memory effects found in RF power amplifiers
  • Use and understand the mathematical algorithms for behavioral modeling and digital pre-distortion
  • Use and implement digital pre-distortion methods for linearization of RF Pas
  • Evaluate and compare modeling and DPD techniques

Mathworks MATLAB is used for all of the mathematical modeling, and Keysight ADS for any circuit simulations for verification, presented in the course. The course is intended to use these tools as a way of showing strategies and examples during the course. It is not mandatory that participants have previous knowledge of using these specific tools.


This course is suitable for:

  • RF PA designers
  • DSP designers
  • System-Level engineers

who are involved in the specification, design, and implementation of linearized RF PAs and transmitter systems, or who are developing pre-distortion methods, software and algorithms for linearization of RF PAs.

Electrical Engineering degree or equivalent and at least two years applicable practical experience is recommended.

Behavioral Modeling and Digital Predistortion of RF Power Amplifiers

Day 1

RF Power Amplifiers in Modern Wireless Communications Systems

  • Modern Wireless Communications modulations formats: LTE, WCDMA
  • Signal Metrics: Peak-to-Average Power Ratio
  • High-efficiency PA architectures: Doherty, Envelope Tracking
  • PA metrics: AM-AM, AM-PM, ACLR, EVM

Introduction to Models and Modeling

  • Compact models, Behavioral models
  • Introduction to System Identification methods

Day 2

Behavioral Modeling Methods

  • Linear System review
  • Nonlinear modeling
  • Memory effects
  • Nonlinear dynamical models: Volterra Series formulations and practical implementations
  • Artificial Neural Networks
  • Frequency-domain models and methods: X-parameters

Day 3

Digital Pre-Distortion Methods

  • Typical architectures and approaches using digital control techniques; 'indirect learning'; LUT and algorithmic approaches
  • Adaption techniques: convergence, optimization
  • DPD implementations: hardware, system components, power and cost
  • Practical and commercial DPD systems overview

Future Developments

  • Carrier aggregation
  • Wide bandwidths
  • SDR
  • etc