The Seventh International Symposium on Wireless Communication Systems

York, The University of York, United Kingdom, 19th-22nd September, 2010

Yves Louët

Recent advances in Peak to Average Power Ratio reduction methods for multi-carrier systems

Prof. Yves Louët
SUPELEC

Motivation and Objectives

Orthogonal Frequency Division Multiplexing (OFDM) transmission has overwhelming advantages as it offers excellent immunity against fading and inter-symbol interferences for high date rate transmission in severe propagation conditions. Nevertheless OFDM signals demonstrate high fluctuations termed as Peak to Average Power Ratio (PAPR), what is viewed as a salient shortcoming and has triggered a huge amount of work since a dozen of years. The reason why PAPR has drawn so much attention is mainly due to power amplification. A signal with high PAPR when amplified by a Power Amplifier (PA) without any special treatment results in severe impairments, the well known non-linear distortions because PAs exhibit non linear characteristics. Hence turns up the problem of efficient power amplification of large envelope signals. First basic precautionary measurement includes PA operation in its linear region such that the peaks of the signal to be amplified never enter into non linear amplifier zone. Although this way of doing nullify non linear distortions it leads to large consumption penalty due to low PA efficiency in linear region. Therefore the amplifier must be linearized so that the operating point stays in the high efficiency zone without going into the saturation region. As a consequence second precautionary measurement is to process the signal to be amplified which means to reduce its fluctuations. Precisely speaking this implies PAPR reduction so as to drive the signal as close as possible to the PA saturation where efficiency is high. The consequence is that output power and coverage are increased what is of utmost importance in broadcast or mobile contexts.

Hence PAPR reduction methods have surfaced up because most of telecommunications standards have embraced OFDM modulation in their physical layer. Viewed as a power saving solution, PAPR reduction has became a hot topic today in telecommunications.

In parallel, with the rapidly growing developments in the field of telecommunications, many applications have surfaced up that require very high data rates. To satisfy the needs of these applications, many protocols and standards have been developed. And with that the need of a universal mobile system is more than ever. Software Radio (SWR) system has the answer to this mingling standards scenario. SWR signal consists of multiple standards where each standard has its own system parameters. Some standards use multi-carrier modulation like OFDM based systems whereas the other use single carrier modulation like Global System for Mobile Communications (GSM). As SWR signal is a combination of these different single carrier and multicarrier modulation based standards thus might inherit high PAPR.

Substantial work has already been done on OFDM PAPR problem. Very little work has yet been done on Software Radio (SWR) signal's PAPR problem. The purpose of this tutorial is to draw a state of the art of PAPR regarding current and emerging telecommunication technologies.

In this context this tutorial's objectives are to answer to the following questions:

  • why PAPR is of such high interest in telecommunications and how is it defined and calculated?
  • what are the methods dedicated to PAPR mitigation?
  • what are the parameters any PAPR mitigation method has to trade off?
  • how can all these methods be classified?
  • how to select a PAPR method in a given context?
  • what are the methods of high interest today for PAPR reduction in future telecommunication standards?
  • what is the PAPR of Software Radio and Cognitive Radio systems?
  • what is the consequence of Dynamic Spectrum Access on PAPR of Cognitive Radio systems?

Abstract

This tutorial aims to overview the salient deficiency of Orthogonal Frequency Division Multiplexing (OFDM) modulation namely its high Peak to Average Power Ratio (PAPR). Firstly it aims to describe PAPR and secondly to propose an original classification of all well known mitigation methods. The last part of this tutorial details very promising methods (like Tone Reservation, Active Constellation Extension, etc.) which are about to be normalized in further telecommunication systems like digital video broadcasting, mobile telecommunications and wireless networks. Moving on from OFDM PAPR analysis, this tutorial is ended by an extension to multistandard signals scenario under the umbrella of emerging Software Radio and Cognitive Radio technologies.

Detailed Outline

The outline of the talk is the following:

  • Part I: PAPR fundamentals
    • I-1 Definitions
    • I-2 Theoretical expressions and upper bounds in OFDM context
    • I-3 PAPR regarding non linear power amplification
  • Part II: PAPR mitigation methods in OFDM context
    • II-1 Compromises to take into account
    • II-2 Classification of methods
    • II-3 Performances
  • Part III: PAPR today and tomorrow
    • III-1 Current PAPR methods applied in telecommunications standards
    • III-2 PAPR problem in Software radio (definition and methods)
    • III-3 PAPR viewed as a sensor for Dynamic Spectrum Access in Cognitive Radio systems

Some Recent Instructor References in this Subject

  • Yves Louët and Jacques Palicot "A classification of methods for efficient power amplification of signals" Annals of Telecommunications Journal, vol. 63, Issue 7-8, pp 351-368, July/August 2008
  • Yves Louët, Jacques Palicot, Sajjad Hussain "Power Amplification issues related to Dynamic Spectrum Access in the Cognitive Radio Systems", Book Chapter, Cognitive Radio Systems ISBN 978-953-7619-25-1, In-Tech, 2009.
  • Jacques Palicot, Yves Louët, Sajjad Hussain, Sidkieta Zabre "Frequency Domain Interpretation of Power Ratio Metric for Cognitive Radio Systems", IET Communications Journal, vol.2, Issue 6, pp. 783-793, July 2008
  • Basel Rihawi and Yves Louët "PAPR Reduction Scheme with SOCP for MIMO-OFDM systems" International Journal of Communications, Network and System Sciences, vol.1, no 1, pp 29-35, Feb. 2008
  • Irène Mahafeno, Yves Louët, Jean-François Hélard " Peak-to-average power ratio reduction using second order cone programming based tone reservation for terrestrial digital video broadcasting systems", IET Communications Journal, vol. 3, Issue 7, pp. 1250-1261, July 2009
  • Yves Louët, Irène Mahafeno, Jean-François Hélard "SOCP approach for PAPR reduction using Tone Reservation for the future DVB-T/H standards" 7th IEEE International Workshop on Multi Carrier and Solution (MC-SS), Herrsching, Germany, May 2009.
  • Sajjad Hussain and Yves Louët "PAPR reduction of Software Radio signals using PRC method", IEEE Sarnoff Symposium, Princeton, NJ, Mar. 2009.
  • Yves Louët, Sajjad Hussain "Peak to Mean Power Ratio Statistical Analysis of continuous OFDM signals" IEEE VTC 08, Singapor, May 08
  • Jacques Palicot, Yves Louët "Power Ratio definitions and analysis in single carrier modulation", EUSIPCO 05, Antalya, Turkey, September 05.

Biography

Yves Louët was born in Quimperlé (France) in 1973. He is an engineer of the ISITV (Institute of Science Engineering), Toulon, France, 1996. He received his M.Sc. in Signal Processing from ENSPM (High National School of Physics), Marseille France, 1996 and his Ph.D. degree in Digital Communications in 2000 from Rennes 1 University, Rennes, France. The subject of his Ph.D. was about OFDM systems and PAPR with channel coding. After, he worked two years in SIRADEL, Rennes, France as expert in digital communications for radio channel propagation modelling. He is now Professor in SUPELEC, Rennes, France and his research activities concern PAPR mitigation methods and Software Radio in SCEE (Signal Communication and Embedded Electronic) lab. of SUPELEC.

History of the Tutorial Presentation

  • IEEE ISWCS 2008 (Reykjavik, Iceland)
  • IEEE ISSPIT 2007 (Cairo, Egypt)

This tutorial has always been updated according to new advances in PAPR issue. This year new results on Software Radio and Cognitive Radio systems are included.