MIMO Communication for Cellular Networks

Howard Huang Constantinos B. Papadias Sivarama Venkatesan

MIMO Communication for Cellular Networks Howard Huang Constantinos B. Papadias Bell Labs, Alcatel-Lucent Athens Information Technology (AIT) 791 Holmdel Road Markopoulo Avenue 19.5 km Holmdel, New Jersey 07733 190 02 Peania, Athens USA Greece [email protected] [email protected]

Sivarama Venkatesan Bell Labs, Alcatel-Lucent 791 Holmdel Road Holmdel, New Jersey 07733 USA [email protected]

ISBN 978-0-387-77521-0 e-I SBN 978-0-387-77523-4 DOI 10.1007/978-0 - 387 - 77523 - 4 Springer New York Dordrecht Heidelberg London

Library of Congress Control Number: 2011942318

© Springer Science+Business Media, LLC 2012 All rights reserved. This work may not be translated or copied in whole or in part without the written permission 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 in connection with any form of information storage and retrieval, electronic adaptation, computer soft- ware, 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 if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com) For Michelle, H.H.

For Maria-Anna, C.B.P.

For my parents, S.V.

Preface

Since the 1980s, commercial cellular networks have evolved over several gen- erations, from providing simple voice telephony service to supporting a wide range of other applications (such as text messaging, web browsing, streaming media, social networking, video calling, and machine-to-machine communi- cation) that are accessed through a variety of devices (such as smart phones, laptops, tablet devices, and wireless sensors). These developments have fueled a demand for higher spectral efficiency so that the limited spectral resources allocated for cellular networks can be utilized more effectively. In parallel, starting in the mid-1990s [1], multiple-input multiple-output (MIMO) wireless communication has emerged as one of the most fertile ar- eas of research in information and communication theory. The fundamental results of this research show that MIMO techniques have enormous potential to improve the spectral efficiency of wireless links and systems. These tech- niques have already attracted considerable attention in the cellular world, where simple MIMO techniques are already appearing in commercial prod- ucts and standards, and more sophisticated ones are actively being pursued.

Goals of the book In this book, we hope to connect these two worlds of MIMO communication theory and cellular network design with the goal of understanding how multi- ple antennas can best be used to improve the physical-layer performance of a cellular system. We attempt to strike a balance between fundamental theoret- ical results, practical techniques and core insights regarding the performance limits of multiple antennas in multiuser networks. Unlike books that focus on the theoretical performance of abstract MIMO channels, this one emphasizes the practical performance of realistic MIMO systems.

vii viii Preface

We present in the first part of the book a systematic description of MIMO capacity and capacity-achieving techniques for different classes of multiple- antenna channels. The second part of the book describes a framework for MIMO system design that accounts for the essential physical-layer features of practical cellular networks. By applying the information-theoretic capacity results to this framework, we present a unified set of system simulation stud- ies that highlight relative performance gains of different MIMO techniques and provides insights into how best to utilize multiple antennas in cellular networks under various conditions. Characterizations of the system-level per- formance are provided with sufficient generality that the underlying concepts can be applied to a wide range of wireless systems, including those based on cellular standards such as LTE, LTE-Advanced, WiMAX, and WiMAX2.

Intended audience The book is intended for graduate students, researchers, and practicing engi- neers interested in the physical-layer design of contemporary wireless systems. The material is presented assuming the reader is comfortable with linear alge- bra, probability theory, random processes, and basic digital communication theory. Familiarity with wireless communication and information theory is helpful but not required.

Acknowledgements We have attempted to represent in this book a small sliver of knowledge accumulated by a vast community of researchers. Over the years, we have had the great pleasure of learning from and interacting with many mem- bers of this community within Bell Labs, Alcatel-Lucent, and at other cor- porations and academic institutions. These experts include Angela Alex- iou, Alexei Ashikhmin, Dan Avidor, Matthew Baker, Krishna Balachan- dran, Liyu Cai, Len Cimini, David Goodman, I˜naki Esnaola,Vinko Erceg, Rodolfo Feick, Jerry Foschini, Mike Gans, David Gesbert, Maxime Guillaud, Bert Hochwald, Syed Jafar, Nihar Jindal, Volker Jungnickel, Kemal Karakay- ali, Achilles Kogiantis, Alex Kuzminskiy, Persa Kyritsi, Angel Lozano, Mike MacDonald, Narayan Mandayam, Laurence Mailaender, Thomas Marzetta, Thomas Michel, Pantelis Monogioudis, Francis Mullany, Marty Meyers, Ar- ogyaswami Paulraj, Farrokh Rashid-Farrokhi, Niranjay Ravindran, Gee Rit- tenhouse, Dragan Samardzija, James Seymour, Sana Sfar, Steve Simon, Tod Sizer, John Smee, Max Solondz, Robert Soni, Aleksandr Stoylar, Said Tatesh, Stephan ten Brink, Lars Thiele, Filippo Tosato, Cuong Tran, Matteo Trivel- Preface ix lato, Giovanni Vannucci, Sergio Verd´u, Harish Viswanathan, Sue Walker, Carl Weaver, Thorsten Wild, Stephen Wilkus, Peter Wolniansky, Greg Wright, Gerhard Wunder, Hao Xu, Hongwei Yang, Roy Yates, and Mike Zierdt. Special thanks go to Antonia Tulino who graciously illuminated various aspects of information theory at a moment’s notice. We would also like to specifically acknowledge the following colleagues who provided valuable feed- back on an earlier draft: George Alexandropoulos, Federico Boccardi, Dmitry Chizhik, Jonathan Ling, Mohammad Ali Maddah-Ali, Chris Ng, Stelios Pa- paharalabos, Xiaohu Shang, and Marcos Tavares. We would like to express our gratitude to Francesca Simkin for her keen eye and expert skill in copy editing the manuscript and to Kimberly Howie for providing tips on improv- ing the visual design. Allison Michael and Alex Greene at Springer provided valuable assistance in the production of the manuscript. Finally, we would like to acknowledge our colleague Reinaldo Valenzuela whose enthusiastic leadership has helped shape the spirit and goals of this book.

Howard Huang and Sivarama Venkatesan Bell Labs, Alcatel-Lucent Holmdel, New Jersey

Constantinos B. Papadias Athens Information Technology Athens, Greece

August 2011

Notation

COL SU-MIMO open-loop capacity C¯OL SU-MIMO average open-loop capacity CCL SU-MIMO closed-loop capacity C¯CL SU-MIMO average closed-loop capacity

CMAC Multiple-access channel capacity region CMAC Multiple-access channel sum-capacity C¯MAC Multiple-access channel average sum-capacity CBC Broadcast channel capacity region CBC Broadcast channel sum-capacity C¯BC Broadcast channel average sum capacity CTDMA TDMA channel maximum achievable sum rate C¯TDMA TDMA channel average maximum achievable sum rate

M SU-MIMO channel: number of transmit antennas MU-MIMO MAC: number of receive antennas MU-MIMO BC: number of transmit antennas Cellular system: number of antennas per base N SU-MIMO channel: number of receive antennas MU-MIMO MAC: number of transmit antennas per user MU-MIMO BC: number of receive antennas per user Cellular system: number of antennas per user K Number of users per base B Number of bases S Number of sectors per site

H, h Complex-valued channel matrix, vector s Transmitted signal vector Q Transmitted signal covariance x Received signal vector G, g Precoding matrix, vector

xi xii Notation n Received noise vector u Vector of data symbols

P Signal power constraint σ2 Noise variance P/σ2 SU- and MU-MIMO channel: average SNR Cellular system: reference SNR 2 H λmax(H) Maximum eigenvalue of HH v Symbol power q Quality-of-service weight α2 Average channel gain d Distance dref Reference distance Z Shadow fading realization G Directional antenna response γ Pathloss coefficient Γ Geometry

E(x) Expected value of random variable x AH Hermitian transpose of matrix A tr A Trace of square matrix A diagA Diagonal elements of square matrix A diag(a1,...,aN )SquareN × N matrix with diagonal elements a1,...,aN

IN N × N identity matrix

0N N × 1 vector of zeroes

C Set of complex numbers R Set of real numbers B Set of precoding matrices U Set of active users Contents

Preface ...... vii

Notation ...... xi

1 Introduction ...... 1 1.1 Overview of MIMO fundamentals ...... 2 1.1.1 MIMOchannelmodels...... 2 1.1.2 Single-usercapacitymetrics...... 6 1.1.3 Multiusercapacitymetrics...... 12 1.1.4 MIMOperformancegains ...... 17 1.2 Overviewofcellularnetworks...... 19 1.2.1 Systemcharacteristics...... 19 1.2.2 Co-channelinterference ...... 26 1.3 Overviewofthebook...... 31

2 Single-user MIMO ...... 35 2.1 Channelmodel...... 35 2.1.1 Analyticalchannelmodels...... 37 2.1.2 Physicalchannelmodels...... 41 2.1.3 Otherextensions...... 42 2.2 Single-userMIMOcapacity...... 43 2.2.1 Capacityforfixedchannels...... 43 2.2.2 Performancegains...... 48 2.2.3 Performancecomparisons...... 53 2.3 Transceiver techniques ...... 54 2.3.1 Linear receivers ...... 55 2.3.2 MMSE-SIC...... 61 2.3.3 V-BLAST...... 63

xiii xiv Contents

2.3.4 D-BLAST...... 66 2.3.5 Closed-loopMIMO...... 68 2.3.6 Space-timecoding...... 68 2.3.7 Codebookprecoding...... 70 2.4 Practicalconsiderations...... 76 2.4.1 CSIestimation...... 76 2.4.2 Spatialrichness...... 77 2.5 Summary...... 77

3 Multiuser MIMO ...... 79 3.1 Channelmodels...... 79 3.2 Multiple-access channel (MAC) capacity region ...... 81 3.2.1 Single-antenna transmitters (N =1)...... 82 3.2.2 Multiple-antenna transmitters (N>1)...... 85 3.3 Broadcastchannel(BC)capacityregion...... 87 3.3.1 Single-antenna transmitters (M =1)...... 89 3.3.2 Multiple-antenna transmitters (M>1) ...... 91 3.4 MAC-BCDuality...... 93 3.5 Scalarperformancemetrics...... 95 3.5.1 MaximizingtheMACweightedsumrate...... 96 3.5.2 MaximizingtheBCweightedsumrate...... 99 3.6 Sum-rateperformance...... 101 3.6.1 MU-MIMO sum-rate capacity and SU-MIMO capacity 103 3.6.2 SumrateversusSNR...... 104 3.6.3 Sum rate versus the number of base antennas M ..... 109 3.6.4 Sum rate versus the number of users K ...... 113 3.7 Practicalconsiderations...... 115 3.7.1 Successive interference cancellation ...... 117 3.7.2 Dirtypapercoding...... 118 3.7.3 Spatialrichness...... 118 3.8 Summary...... 119

4 Suboptimal Multiuser MIMO Techniques ...... 121 4.1 Suboptimal techniques for the multiple-access channel ...... 121 4.1.1 Beamformingforthecaseofmanyusers...... 122 4.1.2 AlternativesforMMSE-SICdetection...... 124 4.2 Suboptimaltechniquesforthebroadcastchannel...... 125 4.2.1 CSITprecoding...... 128 Contents xv

4.2.2 CDITprecoding...... 141 4.2.3 Codebookprecoding...... 142 4.2.4 Randomprecoding ...... 146 4.3 MAC-BC duality for linear transceivers ...... 149 4.3.1 MACpowercontrolproblem...... 151 4.3.2 BCpowercontrolproblem...... 152 4.4 Practicalconsiderations...... 153 4.4.1 ObtainingCSITinFDDandTDDsystems...... 153 4.4.2 Referencesignalsforchannelestimation...... 154 4.5 Summary...... 156

5 Cellular Networks: System Model and Methodology ...... 159 5.1 Cellularnetworkdesigncomponents ...... 160 5.2 Systemmodel...... 164 5.2.1 Sectorization...... 167 5.2.2 ReferenceSNR...... 169 5.2.3 Cellwraparound ...... 174 5.3 Modesofbasestationoperation...... 176 5.3.1 Independentbases...... 178 5.3.2 Coordinatedbases...... 181 5.4 Simulationmethodology...... 186 5.4.1 Performancecriteria...... 188 5.4.2 Iterativeschedulingalgorithm ...... 190 5.4.3 Methodologyforproportional-faircriterion ...... 194 5.4.4 Methodologyforequal-ratecriterion...... 202 5.5 Practicalconsiderations...... 213 5.5.1 Antennaarrayarchitectures ...... 213 5.5.2 Sectorization...... 215 5.5.3 SignalingtosupportMIMO ...... 218 5.5.4 Schedulingforpacket-switchednetworks...... 219 5.5.5 AcquiringCQI,CSIT,andPMI...... 222 5.5.6 Coordinatedbasestations...... 223 5.6 Summary...... 225

6 Cellular Networks: Performance and Design Strategies ... 227 6.1 Simulationassumptions...... 227 6.2 Isolatedcell...... 229 6.3 Cellular network with independent single-antenna bases . . . . . 232 xvi Contents

6.3.1 Universal frequency reuse, single sector per site ...... 233 6.3.2 Throughput per area versus cell size ...... 236 6.3.3 Reducedfrequencyreuseperformance...... 238 6.3.4 Sectorization...... 243 6.4 Cellular network with independent multiple-antenna bases . . . 249 6.4.1 Throughput scaling ...... 249 6.4.2 Fixed number of antennas per site ...... 251 6.4.3 Directionalbeamforming...... 263 6.4.4 IncreasingrangewithMIMO ...... 267 6.5 Cellularnetwork,coordinatedbases...... 269 6.5.1 Coordinatedprecoding...... 269 6.5.2 NetworkMIMO...... 272 6.6 Practicalconsiderations...... 276 6.7 Cellularnetworkdesignstrategies ...... 280

7 MIMO in Cellular Standards ...... 285 7.1 UMTS...... 286 7.2 LTE...... 288 7.3 LTE-Advanced...... 293 7.4 IEEE 802.16e and IEEE 802.16m ...... 295

References ...... 297

Index ...... 309