Optimizing Wireless Communication Systems

Francisco Rodrigo Porto Cavalcanti ·Soren AnderssonEditors

Optimizing WirelessCommunication Systems

123

EditorsFrancisco Rodrigo Porto CavalcantiUniversidade Federal do CearaFortaleza-CECampus do Pici, Bloco 910Brazilrodrigo@gtel.ufc.br

Soren AnderssonEricsson ABIsafjordsgatan 14ESE-164 80 StockholmSwedensoren.n.andersson@ericsson.com

ISBN 978-1-4419-0154-5 e-ISBN 978-1-4419-0155-2DOI 10.1007/978-1-4419-0155-2Springer Dordrecht Heidelberg London New York

Library of Congress Control Number: 2009931759

c© Springer Science+Business Media, LLC 2009All rights reserved. This work may not be translated or copied in whole or in part without the writtenpermission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York,NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use inconnection with any form of information storage and retrieval, electronic adaptation, computersoftware, or by similar or dissimilar methodology now known or hereafter developed is forbidden.The use in this publication of trade names, trademarks, service marks, and similar terms, even ifthey are not identified as such, is not to be taken as an expression of opinion as to whether or notthey are subject to proprietary rights.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

“To Eduarda and Renesa”

Foreword

In June 2000, GTEL (Wireless Telecommunications Research Group) at the Fed-eral University of Ceara was founded by Professor Rodrigo Cavalcanti and his col-leagues with the mission of developing wireless communications technology andimpact the development of the Brazilian telecommunications sector. From the start,this research effort has been supported by Ericsson Research providing a dynamicenvironment where academia and industry together can address timely and relevantresearch challenges. This book summarized much of the research output that hasresulted from GTEL’s efforts. It provides a comprehensive treatment of the physicaland multiple access layers in mobile communication systems describing differentgenerations of systems but with a focus on 3G systems. The team of Professor Cav-alcanti has contributed scientifically to the development of this field and built up animpressive expertise. In the chapters that follow, they share their views and knowl-edge on the underlying principles and technical trade-offs when designing the airinterface of 3G systems.

The complexity of 3G systems and the interaction between the physical and mul-tiple access layers present a tremendous challenge when modeling, designing, andanalyzing the mobile communication system. Herein, the authors tackle this prob-lem in an impressive manner. Their work is very much in line with the developmentsin 3GPP providing a deeper understanding of the evolution of 3G and also futureenhancements. Two main themes are treated, resource management and transceiverdesigns. A common thread in both themes is the use of multi antenna systems orMIMO systems to enhance system performance.

Researchers or engineers active in wireless communications and interested in thedesign and optimization of current and emerging mobile systems are encouraged toshare the results and insights of this comprehensive book.

KTH - Royal Institute of Technology Professor Bjorn OtterstenStockholm, Sweden

vii

Preface

Introduction

Mobile and wireless communication systems are a prominent communications tech-nology with profound economical and social impacts in practically all parts of theworld. The current state of wireless communication systems allows for a much widerscope of applications than what it used to be originally, that is, to be a mobile exten-sion of the public switched telephone network. The convergence of mobile systemsand the Internet has become a reality as new radio access technologies emerged withimproved coverage, capacity, and latency. While the desire to develop and establisha truly mobile Internet dates back to the mid-1990s, it is only now that a signifi-cant increase in the volume of data is being witnessed by most cellular operators,not only in Europe and Japan, but also throughout North and Latin Americas. Thisbook is about some of the underlying technological breakthroughs that allowed theevolution to the current state of development in wireless technology.

The focus of the book is on the two lower layers of the ISO/OSI layered model,that is, the physical and data link layers, including the link and media access con-trol sublayers. These two layers are of specific importance in wireless systems, asopposed to many of its wired counterparts. This is fundamentally due to spectrumshortage, the broadcast nature of interference, and time variability of the wirelesschannel. As a consequence, much of the improvements in coverage, capacity, andlatency of modern wireless systems are due to new approaches for tackling old prob-lems in high-capacity radio communications in these two lower layers.

Intended Audience and Usage

This book is intended for researchers in the field of wireless communications,more specifically to the ones involved with the design and optimization of currentand emerging wireless access technologies for mobile communications. Graduatestudents working in subjects such as radio resource management, OFDM, andMIMO, as well as in third-generation systems and beyond, will benefit from the

ix

x Preface

state-of-the-art concepts, methods, examples, and case studies presented. Everychapter, in addition to having a clear ambition to address the state of the art ofthe corresponding subject, discusses basic concepts in the introductory sections andgives references for the interested reader to deepen his/her understanding. All chap-ters can be used independently as a complement to a graduate-level “advanced”wireless communications course, where each chapter can be subject to a directedstudy or a seminar. The book may also be of interest to the practitioner or to en-gineers involved in standardization efforts. The attention to technical details fromstandards is given in several chapters when performance results and case studiesare presented. The idea is to demonstrate how advanced concepts can be adaptedto be applicable in more realistic scenarios. Finally, almost every chapter of thebook sheds light, directly or indirectly, on the subject of performance evaluationof wireless systems by means of system and link-level simulations. As the com-plexity of wireless systems grows, efficient and correct methods for modeling andperformance simulations of these systems are becoming a fundamental disciplineon their own.

Organization of the Book

Part I – Resource Allocation

Radio resource allocation (also known as radio resource management or RRM) hasits roots in frequency reuse planning of first-generation cellular systems. Its funda-mental goal is to increase spectrum efficiency. More efficient utilization of the radiospectrum plays such an important role because spectrum is simultaneously a veryscarce and widely shared resource.

In the evolution of second- and third-generation systems, RRM became a dis-cipline on its own, encompassing a variety of techniques such as power control,frequency hopping, dynamic channel allocation, and more advanced multi-antennaconcepts, such as beamforming solutions as well as various transmit diversityschemes. Then, the emergence of packet-switched data services in third-generationsystems and beyond has demanded a new set of RRM techniques able to handlemixed services scenarios. These included concepts borrowed from wired data net-works, such as packet scheduling and congestion control, but that were reformulatedand adapted to the wireless environment. More recently, highly configurable emerg-ing radio access technologies, such as orthogonal frequency division multiplexing(OFDM)-based multiple access, have widened the scope of RRM. By means of ad-vanced optimization approaches, radio resource allocation in time and frequency isnow possible with fine granularity, increasing the efficiency potential of spectrumusage to unprecedented levels. This is mainly due to a clever exploitation of themultiuser diversity made available by these emerging systems.

Chapter 1 deals with power control. Transmission power is one fundamental re-source whose optimization impacts directly on coverage and capacity. Power control

Preface xi

has been a key technique since second-generation systems to achieve energy effi-ciency and interference management. This chapter focuses mostly on the latter. Abasic introduction to some fundamentals in wireless communications is included.Basic propagation phenomena and modeling are first discussed. Then, a reviewabout the fundamentals of power control is given along with classical algorithms,including analysis of convergence. A new approach to power control, based on gametheory, is then presented, appropriated to emerging systems where multiple serviceswith different quality-of-service (QoS) demands coexist. In particular a class of “op-portunistic” distributed power control algorithms is derived for elastic data services,making it relevant to reconsider the supremacy of adaptive modulation and codingin current wireless systems. Finally a discussion about the use of channel predictionmethods to improve the performance of existing algorithms is presented.

Chapter 2 presents an overview of RRM for the commercially most successfulmobile communication system to date, that is, GSM, along with its packet-switchedcounterpart, EDGE. RRM has played a key role in the long-lasting life of GSM,which, dating back to more than 25 years now, is still able to cope with the majorityof worldwide voice traffic. The chapter begins with a review on the fundamentalsof the GSM/EDGE technology according to the respective 3GPP standards. Then,several RRM techniques are described as applied to GSM/EDGE along with perfor-mance results, using a detailed and realistic simulation model. These include powercontrol, dynamic channel allocation, spatial division multiple access (SDMA), andmanagement of multiple services by interference balancing. A discussion aboutlarge-scale modeling and simulation of wireless systems is also presented, includingtraffic modeling of data services.

Chapter 3 is a practitioner-oriented tutorial on HSPA deployment and optimiza-tion. HSPA is the key access technology currently behind the mobile broadbandInternet expansion. The chapter serves a dual scope. First, a review about the HSPAstandard is given. Both HSDPA and HSUPA are presented in aspects such as pro-tocol stack, network architecture, channel structure, and physical layer procedures.A description of radio resource management fundamentals in HSPA is presentedincluding aspects such as power allocation, mobility management, and related pro-tocol aspects. Then the author describes several field results and real case studiesleading to optimized broadband experience via HSPA. The chapter ends with sug-gestive guidelines for planning and dimensioning HSPA networks for the residentialmarket.

Chapter 4 builds on the previous chapter to propose and analyze advanced con-gestion control mechanisms for HSPA, as well as for WCDMA (wideband codedivision multiple access) systems. While the baseline WCDMA/HSPA system canbring significant capacity improvements over GSM/EDGE, the growing demandfor data services may rapidly press its spectrum efficiency to the limit. Quality-of-service management by means of congestion control is then proposed for dealingwith multiple services competing for radio resources. Congestion control functions,in the form of admission control, load control, and packet scheduling, are responsi-ble for keeping the network load at controlled levels and maintaining stability whileensuring QoS levels. Basic concepts and new methods are discussed and results

xii Preface

showing the capacity benefits of employing congestion control demonstrate a signif-icant impact. The proposed methods are shown to be fully automatic and scalable,able to cope with many services under different network loads. Case studies forWCDMA and HSPA are presented using realistic simulation scenarios composed ofservices such as World Wide Web access and voice-over-IP.

Chapter 5 addresses state-of-the-art OFDMA systems and corresponding re-source allocation aspects. As previously mentioned, OFDMA opens up a new breedof RRM techniques due to the high flexibility and granularity with which frequencyand time radio resources (i.e., subcarriers and time slots) can be allocated to mul-tiple users. Advanced optimization techniques can then be employed to map radioresources to active connections in such a way as to fulfill network-level objectivessuch as maximization of the overall capacity or satisfaction of QoS levels. The chap-ter begins by establishing the system-level scenarios for RRM in OFDMA and theirdifferences. Then a review of the key optimization and algorithmic approaches suit-able for these problems is given. A new scheduling approach for OFDMA is pro-posed, based on the maximization of the user satisfaction ratio. A case study for3GPP’s long-term evolution (LTE) system is presented to illustrate the performanceof the proposed methods and concepts. Finally, a new method for power alloca-tion for OFDMA is presented along with results showing superior performance ascompared to existing approaches.

Finally (for Part I), Chapter 6 looks to the near future of wireless systems bydealing with the topic of multi-access networks. In this case, multiple radio accesstechnologies cooperate to increase coverage and capacity. By means of a commoncore network infrastructure, complementary features of different radio access tech-nologies can be combined to increase return of investment of existing networkswhile attending new demands for coverage and capacity. The chapter begins with aconceptual review about multi-access networks and the involved fundamental trade-offs. Then, concepts and methods for common radio resource management are ex-posed. These methods can be seen as an extension of conventional RRM methodsfor the multi-access case. Typical CRRM procedures include access selection andinter-system (or vertical) handovers. A case study involving a UMTS (UniversalMobile Telecommunication System) and a wireless local area (WLAN) joint net-work is explained and illustrated with simulation results.

Part II – Transceiver Architectures

The significant improvements at the physical layer have been instrumental for theincrease of the wireless link capacity over the last decade. OFDM itself, already apopular modulation mechanism in fixed digital subscriber lines, has been combinedwith the use of multiple antennas at both ends of wireless links, in the so-calledmultiple input multiple output (MIMO) schemes. MIMO has changed the waywireless engineers face the fundamental capacity limits of the wireless channelby exploiting fading variability in favor of it. This fact also illustrates the major

Preface xiii

challenge – How can a wireless system be designed that allows for a practical im-plementation in the presence of such potentially fast fading propagation channelsbetween and among the multitude of employed antennas? The main aspect to takeinto consideration is how to make such a system design both observable and con-trollable – the former important in order to generate the appropriate amount of radionetwork measurements and the associated signaling and the latter significant in thesense of keeping the interference levels under control on a system level.

The understanding and modeling of MIMO propagation channels have reacheda rather mature level during the last decade; a remaining problem is, however, thecomputational complexity associated with using any of the available detailed MIMOmodels in system (or even link-) level simulations. There are also still many aspectsto understand when it comes to including also antenna design – and modeling as-pects for any realistic MIMO application – this is in particular the case on the userequipment side, mainly due to the fundamental restrictions originating from the size(in wavelengths) of handheld or portable devices.

Chapter 7 deals with a basic concept when analyzing wireless links by explain-ing the way wireless links can be modeled and have their performance efficientlyevaluated. Modeling and simulating wireless systems is a complex task which startswith a good assessment of the physical layer behavior. The chapter discusses twomain aspects. First, the authors discuss the approaches for dividing complex wire-less system simulations into two independent, more tractable parts, namely link andsystem-level simulations. Then they focus on how to design reliable link-level simu-lators. Besides that, a software development framework is proposed for flexible andmodular construction of link-level simulators. Several case studies are presented,involving the modeling and simulation of actual mobile systems, to illustrate theconcepts.

Chapter 8 presents an overview of techniques related to the problem of equal-ization for wireless systems. The hereby desired recovery of coded symbols trans-mitted through a propagation channel is treated for the SISO scenario as well as forthe SIMO (beamforming) case. Techniques for channel identification and trackingare discussed together with means to handle time-dispersive channels using eithertime- or frequency-domain techniques. Furthermore, case studies exemplify typicalequalization solutions for wireless systems in use today. Finally, the chapter dis-cusses the concept of – and principles for – turbo-equalization, that is, equalizationstructures that achieve near-optimal performance by jointly performing equalizationand decoding.

Chapter 9 treats channel estimation for OFDM-based systems. Since the granu-larity in the time and frequency-domains are rather high, the concepts of frequency-domain interpolation and time-domain filtering are required for a practicalimplementation of a channel estimation algorithm. These concepts become veryimportant for allowing as low a density as possible of the reference or pilot signalsdedicated for aiding the channel estimation over the frequency band of interest andover time as the channel changes. As mandated by the OFDM systems currentlyemerging from ongoing standardization efforts (in, e.g., 3GPP LTE), good chan-nel estimation performance and robustness as well as the associated system design

xiv Preface

aspects – like the desire to reduce the signaling overhead – becomes highly relevantfor MIMO applications. Such aspects are discussed in that chapter and differentchannel estimators are evaluated and compared by means of simulations.

Chapter 10 brings the discussions in Chapter 9 further toward an applicationof channel estimation methods and the related channel state information and chan-nel quality indicators to the problem of adapting modulation scheme and codingrate for a MIMO-OFDM system. This is commonly known as the problem of link-adaptation, and the problem to address is how to best select transmission parameters(like transmit antennas, bit rates, transmit power) for a certain estimated channel re-alization in order to optimally utilize the available system resources (like spectralefficiency or ultimately even energy consumption). The chapter outlines and dis-cusses the different gains that can be achieved – diversity and multiplexing – andthe relation between them. Furthermore, some hybrid MIMO transmission schemesare suggested and evaluated for various numbers of employed transmit antennas.

In Chapter 11 the authors present an innovative space–time–frequency multiple-access (STFMA) MIMO wireless communication system combining space-domainand frequency-domain spreading by means of linear precoding, along with a time-domain block-spreading CDMA strategy. Precoding across space (transmit anten-nas) and frequency (subcarriers) provides robustness against deep channel fadeswhile providing space and frequency diversities, while block-spreading enablesmultiple accessing. They utilize a tensorial algebra-based decomposition to modelthe received signal in the STFMA system. Thanks to the powerful identifiabilityproperties of this tensor decomposition, blind signal detection based on multiuserdetection is possible.

Chapter 12 finally addresses the problem of how to reduce the overhead sig-naling that is typically present for MIMO transmit schemes employing closed-loopchannel state information feedback. The techniques that are discussed are mainlybased on the concept of transmitter precoding, that is, the feedback from the userequipment consists of an index in a pre-designed transmit codebook, known at bothtransmitter and receiver sides; the codebook design is also described in the chapterin the form of illustrative examples. The transmitter then applies the so-indicatedcodebook vector of antenna weights, and this process is updated regularly. Clearly,the performance of closed-loop precoding schemes will suffer from high-speed ter-minals since the selected codebook index then quickly becomes outdated, and apossible remedy is then to switch to an open-loop transmit scheme, where mainlythe modulation and coding rates are updated regularly and the potential additionaltransmit weights are designed to primarily generate diversity gains. Examples areprovided, where the performance of different transmit schemes are compared bymeans of simulation studies.

Fortaleza, Brazil Francisco R. P. CavalcantiStockholm, Sweden Soren N. Andersson

Acknowledgments

This book is the result of almost 10 years of research activities at the WirelessTelecommunications Research Group (GTEL) at the Federal University of Ceara(UFC), Brazil. GTEL was created in the year 2000 as a joint effort among UFC,Ericsson Brazil and Ericsson Research in Sweden. The present book – apart fromall the M.Sc. and Ph.D. students that have been produced – can be seen as a directand very successful result of this stimulation effort.

We are particularly thankful to Eduardo Oliva, Maria Valeria Marquezini, andAndrea Barros of Ericsson Brazil, who have managed the strategic, formal, andlegal aspects of the research projects over the years with GTEL, as well as themanagement at Ericsson Brazil, Fernando Aragao and Trond Fidje, who havebeen supportive in the efforts involved in this research cooperation throughout theyears.

We express our gratefulness also to the many students that have taken part in andcontributed to the research efforts over the years, some of them chapter authors inthis very book and turned into professors and industry experts now. Among them wewould like to mention with special gratitude Andre Almeida, Charles Cavalcante,Emanuel Bezerra, Leonardo Sampaio, Tarcisio Maciel, Vicente Souza, WaltemarSousa, Walter Cruz, and Yuri Silva. We would also like to acknowledge professorsJoao Mota and Joao Romano for their various and invaluable contributions to theconsolidation of GTEL since its foundation. Participating chapter authors from otherinstitutions are thanked for their long-term and prolific partnership. One further andspecial thank you is to Mrs. Ana Carvalho for her endless dedication as executivesecretary at GTEL.

Previously at Ericsson Research and currently with Ericsson’s Business Unit Net-works, a special thank you is passed on to Bo Goransson for his feedback over theyears regarding the many detailed research issues of relevance to consider within thearea of multiple antenna systems and to Henrik Asplund in the Propagation Group atEricsson Research for his guidance on radio wave propagation aspects and channelmodeling intricacies.

An expression of particular gratitude finally goes to the current and previousresearch managers Mikael Hook, Sverker Magnusson, and Sven-Olof Jonsson at

xv

xvi Acknowledgments

Ericsson Research who over the years have supported, directed, and reviewed thedifferent projects’ ambitions and results from the GTEL research cooperation.

Fortaleza, Brazil Francisco R. P. CavalcantiStockholm, Sweden Soren N. Andersson

Contents

Part I Resource Allocation

1 Power Control for Wireless Networks: Conventionaland QoS-Flexible Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3F. de S. Chaves, F. R. P. Cavalcanti, R. A. de Oliveira Neto,and R. B. Santos1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.2 Models and Basic Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.3 Centralized Power Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.4 Distributed Power Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111.5 Feasibility and Convergence Aspects of Distributed Power Control 151.6 Power Control for QoS-Flexible Services . . . . . . . . . . . . . . . . . . . . . 181.7 Power Control Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231.8 Prediction of Channel State Information . . . . . . . . . . . . . . . . . . . . . . 411.9 Conclusions and Topics for Future Research . . . . . . . . . . . . . . . . . . . 46References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

2 RRM Performance for GSM/EDGE Radio Acess Network . . . . . . . . . 51Y. C. B. Silva, T. F. Maciel, and F. R. P. Cavalcanti2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 512.2 Fundamentals of RRM in GSM/EDGE . . . . . . . . . . . . . . . . . . . . . . . 522.3 Advanced Radio Resource Management for GSM/EDGE . . . . . . . . 582.4 Simulation and Modeling of GSM/EDGE Networks . . . . . . . . . . . . 652.5 RRM Performance in GSM/EDGE . . . . . . . . . . . . . . . . . . . . . . . . . . . 712.6 Conclusions and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . 91References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

3 Performance Optimization in Practical HSPA Networks forWireless Broadband Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95M. I. J. Da Silva3.1 Introduction to Broadband Wireless Access . . . . . . . . . . . . . . . . . . . 953.2 System Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 963.3 HSDPA Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

xvii

xviii Contents

3.4 HSDPA Field Trials: Mobility Issues . . . . . . . . . . . . . . . . . . . . . . . . . 1183.5 HSUPA Results: Field Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1193.6 Applications Performance over HSPA . . . . . . . . . . . . . . . . . . . . . . . . 1203.7 Capacity Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1343.8 Conclusion and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . 139References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

4 Congestion Control for Wireless Cellular Systems with Applicationsto UMTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141E. B. Rodrigues, F. R. P. Cavalcanti, and S. Wanstedt4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1414.2 Congestion Control and QoS Management . . . . . . . . . . . . . . . . . . . . 1424.3 Congestion Control Framework and Radio Resource Management 1454.4 Resource-Based and QoS-Based Congestion Control . . . . . . . . . . . . 1484.5 Resource-Based Framework for Circuit-Switched Networks . . . . . . 1514.6 Case Study: WCDMA Performance

with Circuit-Switched Voice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1584.7 QoS-Based Framework for Packet-Switched Networks . . . . . . . . . . 1654.8 Case Study: HSDPA Performance with VoIP

and WWW Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1744.9 Conclusions and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . 180References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182

5 Resource Allocation in Multiuser Multicarrier Wireless Systemswith Applications to LTE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187W. Freitas Jr., F. R. M. Lima, R. B. Santos, and F. R. P. Cavalcanti5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1875.2 Scenarios for Radio Resource Allocation . . . . . . . . . . . . . . . . . . . . . . 1895.3 Radio Resource Allocation Fundamental Problems . . . . . . . . . . . . . 1935.4 Optimization Problems in Multicarrier Resource Allocation . . . . . . 1965.5 Optimization Tools for Multicarrier Resource

Allocation Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1995.6 Algorithms for Frequency Resource Assignment . . . . . . . . . . . . . . . 2085.7 Subcarrier Assignment in 3GPP’s Long-Term Evolution (LTE) . . . 2145.8 Power Allocation Algorithms and Performance in OFDMA . . . . . . 2215.9 Conclusions and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . 228References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

6 Common RRM for Multiaccess Wireless Networks . . . . . . . . . . . . . . . 233A. P. da Silva, L. S. Cardoso, V. A. de Sousa Jr., and F. R. P. Cavalcanti6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2336.2 Multiaccess Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2346.3 Common Radio Resource Management . . . . . . . . . . . . . . . . . . . . . . . 2366.4 Performance of Access Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2436.5 Access Selection Solutions Performance in Practical Scenarios . . . 2496.6 Performance of Access Selection and Vertical Handover . . . . . . . . . 254

Contents xix

6.7 Case Study: Access Selection in an UTRAN and WLAN . . . . . . . . 2576.8 Conclusions and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . 261References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263

Part II Transceiver Architectures

7 Strategies for Link-Level Performance Assessment in theSimulation of Wireless Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269E. M. G. Stancanelli, C. H. M. de Lima, and D. C. Moreira7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2697.2 Rationale for Link-Level Performance Evaluation . . . . . . . . . . . . . . 2707.3 Link-Level Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2727.4 Link-Level Software Development Framework . . . . . . . . . . . . . . . . . 2817.5 Design of Link-to-System Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . 2917.6 Conclusions and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . 306References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

8 Channel Equalization Techniques for Wireless CommunicationsSystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311C. M. Panazio, A. O. Neves, R. R. Lopes, and J. M. T. Romano8.1 Introduction and Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3118.2 Channel Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3138.3 Equalization Criteria and Adaptive Algorithms . . . . . . . . . . . . . . . . . 3148.4 Improving Equalization Performance Over Time

Dispersive Channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3248.5 Equalization with Multiple Antennas . . . . . . . . . . . . . . . . . . . . . . . . . 3288.6 Turbo-equalization: Near Optimal Performance in Coded Systems 3368.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351

9 Channel Estimation for OFDM Systems: Techniques, Algorithms,and Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353R. F. Vigelis, D. C. Moreira, and C. C. Cavalcante9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3539.2 OFDM Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3539.3 Channel Estimation for Time-Varying Channels . . . . . . . . . . . . . . . . 3659.4 Recursive Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3769.5 Channel Estimation for MIMO-OFDM Wireless Systems . . . . . . . . 3819.6 Conclusions and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . 387Appendix 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388Appendix 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391

xx Contents

10 Link Adaptation for MIMO-OFDM Systems . . . . . . . . . . . . . . . . . . . . 393D. C. Moreira, W. C. Freitas Jr., C. A. de Araujo, and C. C. Cavalcante10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39310.2 Fundamentals of MIMO Transceiver Architectures . . . . . . . . . . . . . 39410.3 Advanced MIMO Transceiver Architectures . . . . . . . . . . . . . . . . . . . 40310.4 Link Adaptation in Multiple Signal Dimensions . . . . . . . . . . . . . . . . 41010.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417

11 Multiuser MIMO Systems Using STFMA PARAFAC TensorModeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421A. L. F. de Almeida, G. Favier, and J. C. M. Mota11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42111.2 Tensor Decompositions: A New Signal Processing Tool . . . . . . . . . 42411.3 Background on the PARAFAC Tensor Decomposition . . . . . . . . . . . 42511.4 Space–Time–Frequency Multiple-Access MIMO System . . . . . . . . 42811.5 STFMA Performance with Perfect Channel Knowledge . . . . . . . . . 43911.6 PARAFAC Tensor Modeling for the STFMA System . . . . . . . . . . . 44411.7 Blind Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44611.8 Simulation Results with Blind Detection . . . . . . . . . . . . . . . . . . . . . . 45211.9 Conclusions and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . 456References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457

12 MIMO Transceiver Design for Enhanced Performance UnderLimited Feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463I. L. J. da Silva, A. L. F. de Almeida, F. R. P. Cavalcanti, and G. Favier12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46312.2 Background on Limited Feedback-Based MIMO Systems . . . . . . . 46512.3 Channel-Adaptive Limited Feedback Beamforming Techniques . . 47212.4 Linear Precoding for Spatial Multiplexing Systems . . . . . . . . . . . . . 48212.5 Linear Precoding for Space–Time-Coded Systems . . . . . . . . . . . . . . 49112.6 Tensor-Based Space–Time Precoding (TSTP) . . . . . . . . . . . . . . . . . . 49312.7 Conclusions and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . 504References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 509

Contributors

Francisco R. P. Cavalcanti received a D.Sc. degree in electrical engineering fromUniversity of Campinas (UNICAMP), Brazil, in 1999. Upon graduation he joinedthe Federal University of Ceara (UFC) where he is an adjunct professor and holdsthe Wireless Communications Chair at the Teleinformatics Engineering Depart-ment. In 2000, he founded and since then has directed GTEL, a research institutebased in Fortaleza, Brazil, focused on the advancement of wireless telecommunica-tions technologies. He is also a program manager directing a program of researchprojects in wireless communications at GTEL sponsored by the Ericsson Researchand Development Center in Brazil. He has published over 100 conference and jour-nal papers in topics related to radio resource management, cross-layer algorithms,and transceiver architectures for wireless systems and networks. Prof. Cavalcantialso holds a “Leadership and Management” professional certificate from the Mas-sachusetts Institute of Technology.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: rodrigo@gtel.ufc.br

Soren Andersson received an M.Sc. EE and Ph.D. degrees, in Automatic Controlfrom Linkoping Institute of Technology, Sweden, in 1988 and 1992, respectively.During 1993 he was a postdoctoral research associate at Yale University. He thenjoined the Department for Access Technologies and Signal Processing – where re-search in advanced antenna systems for wireless networks was initiated – at EricssonResearch, Ericsson AB, Stockholm, Sweden in 1994. There he was active in re-search on adaptive antennas in cellular systems and was project manager for theresearch and subsequent field-trials carried out with respect to the application ofadaptive antennas for GSM. Between 1998 and 2008 he managed Ericsson Re-search’s activities in the area of antenna systems and propagation, and in 2008 hewas appointed as an expert in multi-antenna systems. His research interests are inthe general areas of statistical signal and array processing for wireless communica-tions and radio access technologies, the hereby implicated product implementationrequirements aspects, as well as radio network issues related to the application ofadvanced antenna solutions for wireless systems.Ericsson AB, Ericsson Research, EAB/TU, Isafjordsgatan 14E, S-164 80,Stockholm, Swedene-mail: soren.n.andersson@ericsson.com

xxi

xxii Contributors

Alex P. da Silva received a B.Sc. in electrical engineering from Federal Universityof Ceara (UFC), Brazil, in 2004. During his graduate studies, he took part in theDouble-Degree Program, receiving, also in 2004, a Generalist Engineer degree fromEcole Centrale de Nantes (ECN) France. He received his M.Sc. degree in teleinfor-matics engineering from UFC, in 2007. Since 2003, he has been working in projectsinside a technical cooperation between UFC and Ericsson of Brazil. These projectsaim at proving solutions for radio resource management for 3G and multi-accessnetworks and radio resource allocation for OFDMA-based systems. He is memberof the Wireless Telecommunications Research Group (GTEL), Fortaleza, Brazil.His research interests include wireless communications and mobile networks, multi-access networks, and OFDMA-based systems.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: alex@gtel.ufc.br

Aline O. Neves received a B.Sc. and an M.Sc. degree in electrical engineering fromthe State University of Campinas (UNICAMP), Brazil, in 1999 and 2001, respec-tively. She received her Ph.D. in 2005, also in electrical engineering, from the Uni-versity Rene Descartes (Paris V), Paris, France. Recently, she has become an as-sistant professor at the Engineering, Modeling and Applied Social Science Centerof the Federal University of ABC, Santo Andre, Brazil. Her research interests con-sist of equalization, channel estimation, source separation, and information theoreticlearning.Centro de Engenharia, Modelagem e Ciencias Sociais Aplicadas, UniversidadeFederal do ABC, Rua Santa Adelia, 166, Santo Andre, SP, Brazile-mail: aline.neves@ufabc.edu.br

Andre L. F. de Almeida received a B.Sc. and an M.Sc. degree in electrical engi-neering from the Federal University of Ceara (UFC), Fortaleza, Brazil, in 2001 and2003, respectively, and the double Ph.D. in sciences and teleinformatics engineer-ing, respectively, from the University of Nice Sophia Antipolis (UNSA), France,and UFC, Brazil, in 2007. In 2002 he was a visiting researcher at Ericsson Re-search, Stockholm, Sweden, where he worked on MIMO channel measurements forindoor propagation modeling. He was a postdoctoral fellow with the I3S labora-tory, CNRS, Sophia Antipolis, France, from January to December 2008. He is nowa senior researcher with the Wireless Telecom Research Group (GTEL), Fortaleza,Brazil, where he has worked in transceiver architectures for wireless systems withinthe GTEL-Ericsson Research cooperation. Dr. Almeida is affiliated with the Depart-ment of Teleinformatics Engineering of the Federal University of Ceara. His mainresearch interests lie in the area of signal processing for communications and includearray processing, blind signal separation and equalization, multiple-antenna tech-niques, multicarrier and multiuser communications. Recent work of Dr. Almeidahas focused on the development of tensor models for transceiver design in wirelesscommunication systems.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: andre@gtel.ufc.br

Contributors xxiii

Carlos H. M. de Lima received a B.Sc. and an M.Sc. degree in electrical engi-neering from the Federal University of Ceara (UFC) in 2002 and 2004, respectively.Since then he has been working as a research scientist. From 2000 to 2005, heworked in the Wireless Telecommunications Research Group (GTEL), Fortaleza,Brazil. In 2005 he was a visiting researcher at Ericsson Research, Luela, Sweden,working on power control techniques for the enhanced uplink HSPA system. In2006, he worked at Nokia Institute of Technology, Manaus, Brazil. Currently, heis pursing his D.Sc. in the Department of Electrical and Information Engineering,University of Oulu, Finland. He is also a member of the research staff of the Centrefor Wireless Communications, Oulu, Finland.Centre for Wireless Communications, University of Oulu, Erkki Koiso-Kanttilankatu 2S-door 90570, Finlande-mail: carlos.lima@ee.oulu.fi

Charles Casimiro Cavalcante received a D.Sc. degree from the University ofCampinas (UNICAMP) in Sao Paulo, Brazil in 2004. Dr. Cavalcante has been work-ing on signal processing strategies for communications where he has several paperspublished and he has worked on funded research projects on the area. He has helda grant for Scientific and Technological Development from the Brazilian ResearchCouncil (CNPq) from 2004 to 2007. Since March 2007 he is a visiting professor atTeleinformatics Engineering Department of UFC and a researcher of the WirelessTelecommunications Research Group (GTEL) where he leads research on signalprocessing and wireless communications. His main research interests are in signalprocessing for communications, blind source separation, wireless communications,and statistical signal processing.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: charles@gtel.ufc.br

Cibelly A. de Araujo received a B.Sc. degree in electrical engineering and a M.Sc.degree in teleinformatics engineering from the Federal University of Ceara (UFC),Fortaleza, Brazil, in 2006 and 2008, respectively. She is currently working towardher D.Sc. degree at the same institution. Since 2005, she has been a researcher atthe Wireless Telecommunications Research Group, Fortaleza, Brazil. Currently, sheis also a researcher within the technical cooperation between GTEL and EricssonResearch. Her research interests include cross-layer aspects for wireless communi-cations, scheduling, link adaptation, and feedback reporting in MIMO-OFDM sys-tems.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: cibelly@gtel.ufc.br

xxiv Contributors

Cristiano Magalhaes Panazio received a B.Sc. and an M.Sc. degree in electricalengineering from the State University of Campinas (UNICAMP), Brazil, in 1999and 2001, respectively. He received his Ph.D. in 2005, also in electrical engineering,from the Conservatoire National des Arts et Metiers (CNAM), Paris, France. In2006, he became assistant professor at Escola Politecnica of the University of SaoPaulo. His research interests include equalization, multicarrier modulation, spreadspectrum techniques, space–time receivers, and synchronization techniques.Laboratory of Communications and Signals, Department of Telecommunicationsand Control, USP, Sao Paulo, Brazile-mail: cpanazio@lcs.poli.usp.br

Darlan C. Moreira received a Bachelor’s degree in electrical engineering and theMaster of Science degree in teleinformatics engineering from the Federal Univer-sity of Ceara (UFC), Brazil, in 2005 and 2007, respectively. He is currently pur-suing the Doctor’s degree at the same institution. He is a member of the WirelessTelecommunications Research Group (GTEL), Fortaleza, Brazil, and since 2004,he has been working in projects within the technical cooperation between GTELand Ericsson Research. In 2007 he was a visiting researcher at Ericsson Research,Stockholm, Sweden, working on channel quality measurement and reporting for3GPP’s long-term Evolution (LTE) wireless system. His research interests includecross-layer aspects of wireless communications, scheduling, and link adaptation inMIMO-OFDM systems.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: darlan@gtel.ufc.br

Elvis M. G. Stancanelli received the B.Sc. degree in Electrical Engineering fromthe State University of Londrina (UEL), Brazil, in 2002. In 2001 and 2002, hetook part in a project under technical cooperation between the University of SaoPaulo (USP), Brazil, and Ericsson Research. In July 2004, he received the M.Sc.degree in Electrical Engineering from the Polytechnic School of the University ofSao Paulo (EPUSP). At the same time, he joined the Wireless TelecommunicationsResearch Group (GTEL), Fortaleza, Brazil, as researcher. Since 2004 he has beenworking in several projects within the technical cooperation between GTEL andEricsson Research, where he developed link-level simulators for wireless standardssuch as GSM/EDGE, WCDMA, HSPA, and 3GPP’s LTE. Currently, he is pursuinghis D.Sc. degree in Teleinformatics Engineering at the Federal University of Ceara(UFC), Brazil. Some of his research interests are interference mitigation, diversitytechniques, system modeling, and applied computational intelligence.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: emiguel@gtel.ufc.br

Contributors xxv

Emanuel Bezerra Rodrigues received B.Sc. and M.Sc. degrees in electrical en-gineering from the Federal University of Ceara (UFC), Brazil, in 2001 and 2004,respectively. He worked in the Wireless Telecom Research Group (GTEL-UFC)from 2001 to 2007 participating in several research projects sponsored by the Er-icsson Research Brazilian Branch. In 2004 he was a visiting researcher at EricssonResearch, Linkoping, Sweden, working on congestion control techniques for thehigh-speed packet access system. He is currently doing his Ph.D. studies at the Sig-nal Theory and Communications Department (TSC) of the Technical University ofCatalonia (UPC), Spain. His main research interests are radio resource management,QoS control, and cross-layer optimization for mobile communication systems.Technical University of Catalonia - UPC, Campus Nord, Jordi Girona 1-3, 08034Barcelona, Spain e-mail: emanuel@tsc.upc.edu

Fabiano de S. Chaves received the B.S. degree in electrical engineering and theM.S. degree in teleinformatics engineering from Federal University of Ceara (UFC),Brazil, in 2003 and 2005, respectively. He was from 2002 to 2005 with the Wire-less Telecommunications Research Group (GTEL), Fortaleza, Brazil. He is con-ducting his doctorate studies at University of Campinas (UNICAMP), Brazil, and isalso member of the IEEE and of the Brazilian Telecommunications Society (SBrT).His research interests include distributed power control for communication systems,non-cooperative game theory, and interplays between signal processing and controlmethodologies.Department of Communications, School of Electrical and Computer EngineeringUniversity of Campinas – UNICAMP, P.O. Box 6101, Campinas, 13083-852, SaoPaulo, Brazile-mail: fabiano@decom.fee.unicamp.br.

Gerard Favier received an engineering diploma from ENSCM (Ecole NationaleSuperieure de Chronometrie et de Micromecanique), Besancon, and ENSAE (EcoleNationale Superieure de l’Aeronautique et de l’Espace), Toulouse, the EngineeringDoctorate and State Doctorate degrees from the University of Nice Sophia Antipolis,in 1973, 1974, 1977, and 1981, respectively. In 1976, he joined the CNRS (CentreNational de la Recherche Scientifique) and now he works as a research directorof CNRS at the I3S Laboratory, in Sophia Antipolis. From 1995 to 1999, he wasthe director of the I3S Laboratory. His present research interests include nonlinearprocess modeling and identification, blind equalization, tensor decompositions, andtensor approaches for wireless communication systems.Laboratoire I3S/UNSA/CNRS, 2000 route des Lucioles, Les Algorithmes/EuclideB BP 121, Sophia Antipolis, Francee-mail: favier@i3s.unice.fr

xxvi Contributors

Icaro L. J. da Silva received a Bachelor degree in electrical engineering and aMaster of science degree in teleinformatics engineering from the Federal Universityof Ceara (UFC), Brazil, in 2006 and 2009 respectively. Currently he is pursuinghis M.Sc. degree in teleinformatics engineering from the same university. Since2006 he has been with the Wireless Telecommunications Research Group (GTEL)where he is a researcher working on MIMO antenna systems and related issues suchas space–time coding, spatial multiplexing and limited feedback. In 2009 he is avisiting researcher at Ericsson Research, Stockholm, Sweden, working in aspectssuch as MIMO precoding and limited feedback.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: icaro@gtel.ufc.br

Joao Cesar M. Mota received his B.Sc. degree in physics from the Federal Uni-versity of Ceara (UFC), Brazil, in 1978, the M.Sc. degree from Pontifıcia Univer-sidade Catolica (PUC-RJ), Brazil, in 1984, and D.Sc. degree from the Universityof Campinas (UNICAMP), Brazil, in 1992, all in telecommunications engineering.Since August 1979, he has been in the UFC, and currently he is professor withthe Teleinformatics Engineering Department. Dr. Mota worked in Institut Nationaldes Telecommunications and Institut de Recherche en Communications et Cyberne-tique de Nantes, both in France, as invited professor during 1996–1998 and spring2006, respectively. He was general chairman of the 19th Brazilian Telecommunica-tions Symposium – SBrT’2001 and the International Symposium on Telecommu-nications – ITS’2006. He is responsible for the international mobility program forengineering students of UFC. His research interests include digital communications,adaptive filter theory, and signal processing. He is member and counselor of the So-ciedade Brasileira de Telecomunicacoes and member of the IEEE communicationsSociety and IEEE Signal Processing Society. He is counselor of the IEEE StudentBranch in UFC.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: mota@gtel.ufc.br

Joao Marcos Travassos Romano received the degrees of engineer and MS in elec-trical engineering from the University of Campinas (UNICAMP), Brazil. He re-ceived his Ph.D. in automatic and signal processing from the University of Paris –XI in 1987. In 1988 he joined, as an associate professor, the School of Electrical andComputer Engineering (FEEC) at UNICAMP where he is currently professor. Since1989, he is recipient of a Research Fellowship from CNPq, the national foundationfor science and technology in Brazil. He has also been an invited professor at theRene Descartes University in Paris and at the Communications and Electronic Lab-oratory in CNAM – Paris. Professor Romano served the Brazilian CommunicationsSociety (SBrT), a sister society of ComSoc-IEEE, as vice president (1996–2000)and president (2000–2004). Professor Romano has over 140 journal and conferencepublications. He has submitted four Brazilian patents and one international patent.He was the advisor of 23 M.Sc. and 12 D.Sc. These is at UNICAMP.Department of Microwaves and Optics, School of Electrical and ComputerEngineering, UNICAMP, PO Box 6101, 13083-852, Campinas, Brazile-mail: romano@dmo.fee.unicamp.br

Contributors xxvii

Leonardo S. Cardoso received an electrical engineering degree from the FederalUniversity of Ceara (UFC), Brazil, in 2003. He received his M.Sc. degree from thesame institution in 2006. From 2001 to 2006 he worked in several projects under atechnical cooperation between UFC and the Brazilian branch of Ericsson Research.Those projects aimed at studying solutions for radio resource management issuesfor 2/3G and multi-access networks. During the same period, he was member of theWireless Telecommunications Research Group (GTEL), Fortaleza, Brazil. In 2006,he joined the Eurecom Institute, France, working in two projects that dealt withmulti-access networks and real-time MIMO channel performance assessment. Healso significantly contributed to the EMOS MIMO platform at the Eurecom insti-tute. He is currently pursuing his Ph.D. at Supelec, France. His research interestsinclude wireless communications, multi-access networks, cognitive radio, and ran-dom matrix theory.Alcatel-Lucent Chair in Flexible Radio - Supelec, 3 rue Joliot-Curie, 91192 Gif-SurYvette Cedex, Francee-mail: leonardo.cardoso@supelec.fr

Mario I. J. Da Silva has been working in telecommunications since 1998. He hasa B.Sc. engineering degree from the Federal University of Ceara (UFC) in Brazil.He also has a Masters degree from the Institut National des Telecommunications inFrance, and during his Masters he joined Motorola Labs in Paris, where he carriedout research on UMTS physical layer performance. He subsequently began workingas a radio design engineer in O2 UK and moved to O2 Ireland in 2001, where heworks as a principal engineer. For the last 7 years, he has been involved in severalprojects on UMTS inclusive of the deployment of broadband over HSPA. He iscurrently working on radio and core optimization.O2 Telefonica Ireland, 28/29 Sir John Rogerson’s Quay, Docklands, Dublin 2,Irelande-mail: mario.dasilva@o2.com

F. Rafael M. Lima received a B.Sc. in electrical engineering and an M.Sc. in tele-informatics engineering from the Federal University of Ceara, UFC, Brazil, in 2005and 2008, respectively. In 2008 he was a visiting researcher at Ericsson Research,Lulea, Sweden, working on packet scheduling techniques and QoS management forthe 3GPPs long-term evolution (LTE) system. He is currently a researcher and aPhD candidate at the Wireless Telecom Research Group, GTEL, working in radioresource allocation for OFDMA-based systems. His research interests include radioresource management to WCDMA/HSDPA networks, packet scheduling, admissioncontrol, link adaptation, and load control.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: rafaelm@gtel.ufc.br

xxviii Contributors

Raimundo Abreu de Oliveira Neto received a Bachelor and Master of Science de-grees in electrical engineering from the Federal University of Ceara (UFC), Brazil,in 2001 and 2004, respectively. From 2002 to 2008 he was with the WirelessTelecommunications Research Group (GTEL) where he has worked as a researcherfor the technical cooperation between GTEL and Ericsson Research. Presently, heis senior engineer at Petrobras - Petroleo Brasileiro S/A. His research interests arepower control, radio resource management, multi-access networks, and economicmodels for telecommunications.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: neto@gtel.ufc.br

Renato da Rocha Lopes received a B.Sc. and an M.Sc. degree in electrical engi-neering from the University of Campinas (UNICAMP), Brazil, in 1995 and 1997,respectively. In 2003, he received the Ph.D. in electrical engineering from the Geor-gia Institute of Technology. Since then, he has been with the School of Electricaland Computer Engineering at UNICAMP, first as a post-doctoral fellow, then, since2006, as an assistant professor. He is the recipient of several scholarships from theBrazilian government. His research interest spans the general area of communi-cations theory, including MIMO systems, turbo receivers, channel estimation andequalization, and multiuser wireless communications.Department of Communications, School of Electrical and Computer Engineering,UNICAMP, PO Box 6101, 13083-852, Campinas, Brazile-mail: rlopes@decom.fee.unicamp.br

Ricardo B. Santos received his B.Sc. in electrical engineering and M.Sc. in tele-informatics engineering from the Federal University of Ceara (UFC), Fortaleza,Brazil, in 2005 and 2008, respectively. Nowadays he is a researcher of Wire-less Telecom Research Group (GTEL) working in radio resource allocation inOFDMA-based systems. His research interests include radio resource managementto WCDMA/HSDPA networks, packet scheduling, admission control, power con-trol, link adaptation, load control, and heuristic algorithms. In 2008 he was a visit-ing researcher at Ericsson Research, Lulea, Sweden, working on packet schedulingtechniques and QoS management for the 3GPP’s long-term evolution (LTE) system.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: brauner@gtel.ufc.br

Contributors xxix

Rui Facundo Vigelis received his B.Sc. degree in electrical engineering in 2005and Master of Science degree in teleinformatics engineering in 2006, both from theFederal University of Ceara (UFC) in Fortaleza, Brazil. Since September 2006 he isworking toward a D.Sc. at UFC in advanced problems of communication systems.He has also worked on funded projects on the subject of OFDM-based wirelesssystem in 2006 where he has published papers on channel estimation methods forwireless systems. His research interests include statistical signal processing, differ-ential geometry, manifold learning, and wireless communications.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: rfvigelis@gtel.ufc.br

Stefan Wanstedt joined Ericsson in 1999 and as a Senior Research Engineer atAdvanced Wireless Ericsson Algorithm Research, Lulea, Sweden. The current fo-cus of his work is on wireless IP optimization, including real-time services overcellular systems, in particular HSPA and LTE. Previous assignments focused on ra-dio network performance measurements for cellular systems, including GPRS andvoice quality models. He has also worked with projects related to streaming overWCDMA and EDGE. He holds a Ph.D. in geophysics from Lulea University ofTechnology.

·

Tarcisio F. Maciel received a B.Sc. and an M.Sc. degree in electrical engineeringfrom the Federal University of Ceara, Fortaleza, Brazil, in 2002 and 2004, respec-tively. He received the Ph.D. in electrical engineering from the Technische Univer-sitat Darmstadt, Darmstadt, Germany, in 2008. In 1999, he attended the TechnischeUniversitat Hamburg-Harburg, Hamburg, Germany, as part of a 1-year sandwichgraduation program. From 2001 to 2004 he was with the Wireless Telecom Re-search Group (GTEL), Fortaleza, Brazil, working in the research projects on radioresource management for wireless systems developed by GTEL in cooperation withEricsson Research. From 2005 to 2008 he was with the Communications Engineer-ing Lab, Darmstadt, Germany, where he developed his Ph.D. studies on resourceallocation for systems with multiple antennas. Currently he is a professor of com-puter engineering at the Federal University of Ceara, Campus of Sobral, and a seniorresearcher at the Wireless Telecommunications Research Group. His main researchinterests are in the areas of wireless communication systems, resource allocation,adaptive antennas, and optimization.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: maciel@gtel.ufc.br

xxx Contributors

Vicente A. de Sousa Jr. received a B.Sc. Electrical Engineer degree from the Fed-eral University of Ceara (UFC), Fortaleza, Brazil, in 2001. During his graduatestudies, he took part in a technical training of Motorola and Eldorado Institute.He received his M.Sc. degree from the UFC, in 2002. Between 2001 and 2006,he had been working in projects inside a technical cooperation between UFC andEricsson of Brazil where developed solutions to smart antennas systems, radio re-source management techniques for 3G networks and interworking of UMTS andWLAN systems. For the same period, he was member of the Wireless Telecom-munications Research Group (GTEL), Fortaleza, Brazil. He is presently workingtoward his D.Sc. degree at UFC, Brazil. Sousa is also R&D coordinator of NokiaTechnology Institute (INdT), Manaus, Brazil. His research interests include wire-less communications and mobile networks, evolutionary computation, multi-accessnetworks, and WiMAX systems.Nokia Institute of Technology (INdT), Rua Torquato tapajos, 7200 - Colonia TerraNova, 69093-415, Manaus, AM, Brazile-mail: vicente.souza@indt.org.br

Walter C. Freitas Jr. received a D.Sc. degree in teleinformatic engineering fromFederal University of Ceara (UFC), Brazil, in 2006 and his B.Sc. and M.Sc. de-grees in electrical engineering from the same university. During his studies, he wassupported by the Brazilian agency FUNCAP and Ericsson. From 2005 to 2006, Dr.Walter was with the Nokia Institute of Technology as a senior researcher. Since 2006he has been a project manager at GTEL working for the Ericsson–GTEL programof projects. From 2008 he is a professor at the Telinformatics Engineering Depart-ment at the Federal University of Ceara. His main area of interest concerns featuresdevelopment to improve the performance of the wireless communication systems,application of link adaptation techniques, OFDMA resource allocation, MIMO sys-tems, and space–time coding.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: walter@gtel.ufc.br

Contributors xxxi

Yuri C. B. Silva received his B.Sc. and M.Sc. degrees from the Federal Universityof Ceara (UFC), Fortaleza, Brazil, in 2002 and 2004, respectively, and the Ph.D.from the Technische Universitat Darmstadt, Germany, in 2008, all in electrical engi-neering. In 1999 he attended the Technische Universitat Berlin, Germany, as part ofa 1-year sandwich graduation program. From 2001 to 2004 he was with the WirelessTelecom Research Group (GTEL), Fortaleza, Brazil, working within the technicalcooperation between GTEL and Ericsson Research. In 2003 he was a visiting re-searcher at Ericsson Research, Stockholm, Sweden, where he developed advancedradio resource management solutions for the GSM/EDGE standard. From 2005 to2008 he was with the Communications Engineering Lab of the Technische Univer-sitat Darmstadt and currently he is a senior researcher at GTEL. His main researchinterests are in the areas of wireless communication systems, resource allocation,adaptive antennas, multicast services, and precoding techniques.Wireless Telecom Research Group (GTEL), Campus do Pici - CP 6005, 60455-760,Fortaleza, CE, Brazile-mail: yuri@gtel.ufc.br

Acronyms

16-QAM 16-Quadrature amplitude modulation3G Third generation3GPP 3rd. Generation Partnership Project4G Fourth generationABC Always best connectedAC Admission controlACK AcknowledgementAcVI Actual value interfaceA-DPCH Associated dedicated physical channelADSL Asymmetric digital subscriber linesALS Alternating least squaresAM Acknowledged modeAMC Adaptive modulation and codingAMR Adaptive multirateAP Access pointARP Allocation/retention priorityARQ Automatic repeat requestAS Access selectionAS Active setASBPC Autonomous SINR balancing power controlATM Asynchronous transfer modeAVI Average value interfaceAWGN Additive white Gaussian noiseBB Branch-and-boundBCCH Broadcast control channelBEP Bit error probabilityBER Bit error rateBGR Benveniste–Goursat–Ruget theoremBLAST Bell Labs layered space–timeBLEP Block error probabilityBLER Block error rate

xxxiii

xxxiv Acronyms

BPSK Binary phase-shift keyingBS Base stationBSC Base station controllerBSS Base station subsystemBTCE Block-type channel estimationBTS Base transceiver stationCAC Call admission controlCC Chase combiningCC Congestion controlCCCH Common control channelCDMA Code-division multiple–accessCESM Capacity ESMCFH Cyclic frequency hoppingCIR Carrier-to-interference ratioCMA Constant modulus algorithmCN Core networkCONFAC Contrained factor decompositionCP Cyclic prefixCPE Customer premise equipmentCPICH Common pilot channelCQ Channel quantizationCQI Channel quality indicatorCRC Cyclic redundancy checkCRESM Cutoff rate ESMCRRM Common radio resource managementCS Circuit-switchedCSE Circuit-switched equivalentCSI Channel state informationCTA Coverage threshold algorithmDBA Distributed balancing algorithmDBLAST Diagonal Bell Labs layered space–timeDCA Dynamic channel allocationDCCH Dedicated control channelDCH Dedicated channelDFE Decision-feedback equalizerDL DownlinkDPC Distributed power controlDPCCH Dedicated physical control channelDPCH Dedicated physical channelDPDCH Dedicated physical data channelDQPSK Differential quadrature phase-shift keyingDS Delay schedulerDS-CDMA Direct-sequence code division multiple accessDTCH Dedicated traffic channelDTX Discontinuous transmission

Acronyms xxxv

E-AGCH E-DCH access grant channelE-DCH Enhanced dedicated channelEDGE Enhanced data rate for GSM evolutionE-DPCCH Enhanced dedicated physical control channelE-DPDCH E-DCH dedicated physical data channelEESM Exponential ESMEFLC Error feedback-based load controlEFR Enhanced full rateEGC Equal gain combiningEGPRS Enhanced general packet radio serviceEGT Equal gain transmissionE-HICH E-DCH hybrid ARQ indicator channeleNB Enhanced Node BEPC Evolved packet coreE-RGCH E-DCH relative grant channelERT Estimated RAN throughput algorithmESM Effective SINR mappingE-TFC E-DCH transport format combinationETSI European Telecommunications Standards InstituteETU Extended typical urbanEUL Enhanced uplinkE-UTRAN Evolved UMTS terrestrial radio access networkFACCH Fast associated control channelFDD Frequency division duplexFDM Frequency division multiplexingFDMA Frequency division multiple accessFEC Forward error correctionFER Frame erasure rateFFT Fast Fourier transformFH Frequency hoppingFIFO First-in-first-outFIFS First-in-first-servedFIR Finite impulse responseFN Frame numberFP Frame ProtocolFSK Frequency shift keyingFSQP Feasible sequential quadratic programmingFSR Frame success rateFTP File Transfer ProtocolG2 Alamouti space–time block code (STBC)G3 3 transmitter antenna STBCGA Genetic algorithmGAP Generalized assignment problemGASP Generalized access selection problemGBR Guaranteed bit rate

xxxvi Acronyms

GERAN GSM/EDGE radio access networkGGSN Gateway GPRS support nodeGMSK Gaussian minimum shift keyingGPRS General packet radio serviceGSM Global system for mobile communicationGW GatewayH-ARQ Hybrid automatic repeat requestHLR Home location registerHMTS Hybrid MIMO transmit schemeHSDPA High-speed downlink packet accessHS-DPCCH High-speed dedicated physical control channelHS-DSCH High-speed downlink shared channelHSN Hopping sequence numberHSPA High-speed packet accessHS-PDSCH High-speed physical downlink shared channelHSUPA High-speed uplink packet accessHTTP Hypertext Transfer ProtocolICI Inter-carrier interferenceIEEE Institute of Electrical and Electronics EngineersIETF Internet Engineering Task ForceIFFT Inverse fast Fourier transformIIR Infinite impulse responseIMS IP multimedia subsystemIMT International Mobile TelecommunicationsIP Internet ProtocolIR Incremental redundancyIRC Interference rejection combiningISI Inter symbol interferenceITU International Telecommunication UnionJLC Jump-based load controlKPI Key performance indicatorKRST Khatri–Rao space–timeL2S Link-to-system-levelLA Link adaptationLAC Link admission controlLBA Load balancing algorithmLC Load controlLD Linear detectionLESM Logarithmic ESMLF Limited feedbackLiESM Linear ESMLL Link-levelLLC Link layer controlLMS Least mean squareLORAF Low-rank adaptive filter

Acronyms xxxvii

LOS Line-of-sightLS Least squaresLSDF Link-Level Software Development FrameworkLTE Long-term evolutionLTI Linear time-invariantLUBA Link utilization balancing algorithmLuT Look-up tableMA Multi-accessMAC Medium access controlMAI Mobile allocation indexMAIO Mobile allocation index offsetMAL Mobile allocation listMANET Mobile ad hoc networkMAP Maximum a posterioriMAT Multi-antenna transmissionMCAS Modulation, coding, and antenna schemeMCBS-CDMA Multi-carrier block-spread code division multiple accessMC-CDMA Multicarrier code division multiple accessMCDS-CDMA Multi-carrier direct-sequence code division multiple accessMCS Modulation and coding schemeMCSE CSE maximization algorithmMC-SSSMA Multi-carrier spread space spectrum multiple accessMIESM Mutual-information ESMMIH Media-independent handoverMIMO Multiple-input multiple-outputMISO Multiple-input single-outputML Maximum likelihoodMLSE Maximum-likelihood sequence estimationMMSE Minimum-mean-square-errorMPF Multicarrier proportional fairMR Maximum rateMR59FR Multi-rate at 5.9 kbit/s with full rateMRC Maximal ratio combiningMRT Maximal ratio transmissionMS Mobile stationMSC Mobile switching centerMSE Mean-squared errorMTDS-CDMA Multi-tone direct sequenceMTSI Multimedia telephony services over IMSMUI Multi-user interferenceMURPA Multiuser residual power allocationNACK Negative acknowledgementNAS Non-access stratumNBAP Node B application partNE Nash equilibrium

xxxviii Acronyms

NLMS Normalized least-mean-square algorithmNRT Non-real timeOF Orthogonality factorOFDM Orthogonal frequency division multiplexingOFDMA Orthogonal frequency division multiple accessOLPC Outer-loop power controlOOP Object-oriented programmingOPC Opportunistic power controlOPC-F Opportunistic power control with fairnessOQ-DPC-1 Opportunistic QoS distributed power control – 1OQ-DPC-2 Opportunistic QoS distributed power control - 2OSI Open systems interconnectionOSIC Ordered successive interference cancellationOVSF Orthogonal variable spreading factorPA Power allocationPACE Pilot-assisted channel estimationPARAFAC Parallel factorPAST Projection approximation subspace trackingPBCCH Packet broadcast control channelPC Power controlPDCP Packet Data Convergence ProtocolPDTCH Packet data traffic channelPDU Protocol data unitPF Proportional fairPhCH Physical channelPHY PhysicalPPC Partial phase combiningPS Packet-switchedPSC Packet schedulingPSK Phase-shift keyingPSTN Public-switched telephone networkQAM Quadrature amplitude modulationQBA Queue-based algorithmQEGT Quantized equal gain transmissionQoS Quality-of-serviceQP Quadratic programmingQPP Quadratic permutation polynomialQPSK Quadrature phase shift keyingQSA Quantized signal adaptationRA Rate adaptationRAN Radio access networkRAT Radio access technologyRB Radio bearerRBER Raw bit error rateRF Radio frequency

Acronyms xxxix

RFH Random frequency hoppingRLC Radio link controlRLS Radio link setRLS Recursive least squaresRM Rate maximizationRMA Rate maximization algorithmRMSE Root mean square errorRNC Radio network controllerRR Round RobinRRA Radio resource allocationRRC Radio resource controlRRM Radio resource managementRT Real-timeRU Resource unitRXLEV Received signal levelRXQUAL Received signal qualitySA Simulated annealingSAC Session admission controlSACCH Slow associated control channelSASP Strict version of the access selection problemSAT Single-antenna transmissionSAW Stop-and-waitSBA Satisfaction balancing algorithmSBPS Service-based power settingSDC Selection diversity combiningSDCCH Stand-alone dedicated control channelSDPC Soft dropping power controlSDT Selection diversity transmissionSDU Service data unitSEA Super exponential algorithmSER Symbol error rateSES Simple exponential smoothingSF Spreading factorSGSN Service GPRS support nodeSHO Soft handoverSIC Successive interference cancellationSIMO Single-input multiple-outputSINR Signal-to-interference-plus-noise ratioSIP Session initiation protocolSIP Signal-interference productSIR Signal-to-interference ratioSISO Single-input single-outputSISO Soft-input/soft-outputSL System-levelSM Spatial multiplexing

xl Acronyms

SMIRA Stepwise maximum-interference removal algorithmSMS Short message serviceSMT Single and multi-antenna transmissionSNDCP Sub-network-Dependent Convergence ProtocolSNR Signal-to-noise ratioSORA Satisfaction-oriented resource allocationSOVA Soft-output Viterbi algorithmSQP Sequential quadratic programmingSRA Stepwise removal algorithmST Space–timeST Subspace trackingSTBC Space–time block codeSTF Space–time–frequencySTFMA Space–time–Frequency Multiple-AccessST-LE Space–time Linear EqualizerSTM Space–time MultiplexingSTS Space–time SpreadingSTTC Space–time Trellis CodeSVD Singular value decompositionTB Transport blockTCH Traffic channelTCP Transport control protocolTDC Time delay compensationTDD Time division duplexTDL Tapped delay lineTDMA Time division multiple accessTF Transport formatTFCI Transport format combination indicatorTPC Transmit power controlTrCH Transport channelTSTP Tensor-based space–time precodingTTI Transmission time intervalUDP User Datagram ProtocolUDPC Up-down power controlUE User equipmentUHPD Users with highest packet delayUL UplinkULA Uniform linear arrayULPT Users with lowest packet throughputULTR Users with lowest transmission rateUM Unacknowledged modeUMTS Universal Mobile Telecommunication SystemURT Estimated user and RAN throughput algorithmUS Uncorrelated scatteringUSF Uplink state flag

Acronyms xli

UTA Utility-based algorithmUTRAN UMTS terrestrial radio access networkVBLAST Vertical Bell Labs layered space–timeVHO Vertical handoverVOFI Variable orthogonality factor interfaceVoIP Voice over IPVQ Vector quantizationVSER Vector symbol error rateWAG WLAN access gatewaysWCDMA Wideband code division multiple-accessWH Walsh–HadamardWIBRO Wireless broadbandWiMAX Worldwide interoperability for microwave accessWLAN Wireless local area networkWPF Weighted proportional fairWSS Wide-sense stationaryWWW World Wide WebZF Zero-forcingZMCSCG Zero mean circularly symmetric complex gaussian