Adaptive Wireless Communications

Transcription

1 Adaptive Wireless Communications Adopting a balanced mix of theory, algorithms, and practical design issues, this comprehensive volume explores cutting-edge applications in adaptive wireless communications, and the implications these techniques have for future wireless network performance. Presenting practical concerns in the context of different strands from information theory, parameter estimation theory, array processing, and wireless communications, the authors present a complete picture of the field. Topics covered include advanced multipleantenna adaptive processing, ad hoc networking, MIMO, MAC protocols, space-time coding, cellular networks, and cognitive radio, with the significance and effects of both internal and external interference a recurrent theme throughout. A broad, self-contained technical introduction to all the necessary mathematics, statistics, estimation theory and information theory is included, and topics are accompanied by a range of engaging end-of-chapter problems. With solutions available online, this is the perfect self-study resource for students of advanced wireless systems, and wireless industry professionals. Daniel W. Bliss is an Associate Professor in the School of Electrical, Computer and Energy Engineering at Arizona State University. Siddhartan Govindasamy is an Assistant Professor of Electrical and Computer Engineering at Franklin W. Olin College of Engineering, Massachusetts.

2 An excellent and well-written book. This book is a must for any wireless PHY system engineer. Vahid Tarokh, Harvard University Great book! Fills a gap in the wireless communication textbook arena with its comprehensive signal-processing focus. It does a nice job of handling the breadth-vs-depth trade-off in a topic-oriented textbook, and is perfect for beginning graduate students or practicing engineers who want the best of both worlds: broad coverage of both old and new topics, combined with mathematical fundamentals and detailed derivations. It provides a great single-reference launching point for readers who want to dive into wireless communications research and development, particularly those involving multi-antenna applications. It will become a standard prerequisite for all my graduate students. A. Lee Swindlehurst, University of California, Irvine

3 Adaptive Wireless Communications MIMO Channels and Networks DANIEL W. BLISS Arizona State University SIDDHARTAN GOVINDASAMY Franklin W. Olin College of Engineering, Massachusetts

4 cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Mexico City Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York Information on this title: / C Dan Bliss and Siddhartan Govindasamy 2013 Dan Bliss s contributions are a work of the United States Government and are not protected by copyright in the United States. This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2013 Printed and bound in the United Kingdom by the MPG Books Group A catalogue record for this publication is available from the British Library Library of Congress Cataloguing in Publication data Bliss, Daniel W., 1966 Adaptive wireless communications : MIMO channels and networks / Daniel W. Bliss, Siddhartan Govindasamy. pages cm Includes bibliographical references and index. ISBN (hardback) 1. MIMO systems. 2. Wireless communication systems. 3. Adaptive signal processing. I. Govindasamy, Siddhartan. II. Title. TK B dc ISBN Hardback Additional resources for this publication at /bliss Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

5 The views expressed are those of the author (D. W. B.) and do not reflect the official policy or position of the Department of Defense or the U.S. Government.

6

7 Contents Preface Acknowledgments page xvii xviii 1 History Development of electromagnetics Early wireless communications Developing communication theory Television broadcast Modern communications advances Early packet-radio networks Wireless local-area networks 10 2 Notational and mathematical preliminaries Notation Table of symbols Scalars Vectors and matrices Vector products Matrix products Norms, traces, and determinants Norm Trace Determinants Matrix decompositions Eigen analysis Eigenvalues of 2 2 Hermitian matrix Singular-value decomposition QR decomposition Matrix subspaces Special matrix forms Element shifted symmetries Eigenvalues of low-rank matrices Matrix inversion Inversion of matrix sum 28

8 viii Contents 2.6 Useful matrix approximations Log determinant of identity plus small-valued matrix Hermitian matrix raised to large power Real derivatives of multivariate expressions Derivative with respect to real vectors Complex derivatives Cauchy Riemann equations Wirtinger calculus for complex variables Multivariate Wirtinger calculus Complex gradient Integration over complex variables Path and contour integrals Volume integrals Fourier transform Useful Fourier relationships Discrete Fourier transform Laplace transform Constrained optimization Equality constraints Inequality constraints Calculus of variations Order of growth notation Special functions Gamma function Hypergeometric series Beta function Lambert W function Bessel functions Error function Gaussian Q-function Marcum Q-function 63 Problems 63 3 Probability and statistics Probability Bayes theorem Change of variables Central moments of a distribution Noncentral moments of a distribution Characteristic function Cumulants of distributions Multivariate probability distributions Gaussian distribution Rayleigh distribution Exponential distribution 73

9 Contents ix Central χ 2 distribution Noncentral χ 2 distribution F distribution Rician distribution Nakagami distribution Poisson distribution Beta distribution Logarithmically normal distribution Sum of random variables Product of Gaussians Convergence of random variables Convergence modes of random variables Relationship between modes of convergence Random processes Wide-sense stationary random processes Action of linear-time-invariant systems on wide-sense stationary random processes White-noise processes Poisson processes Eigenvalue distributions of finite Wishart matrices Asymptotic eigenvalue distributions of Wishart matrices Marcenko Pastur theorem Estimation and detection in additive Gaussian noise Estimation in additive Gaussian noise Detection in additive Gaussian noise Receiver operating characteristics Cramer Rao parameter estimation bound Real parameter formulation Real multivariate Cramer Rao bound Cramer Rao bound for complex parameters 105 Problems Wireless communications fundamentals Communication stack Reference digital radio link Wireless channel Thermal noise Cellular networks Frequency reuse Multiple access in cells Ad hoc wireless networks Achievable data rates in ad hoc wireless networks Sampled signals 137 Problems 138

10 x Contents 5 Simple channels Antennas Line-of-sight attenuation Gain versus effective area Beamwidth Channel capacity Geometric interpretation Mutual information Additive Gaussian noise channel Additive Gaussian noise channel with state Energy per bit 165 Problems Antenna arrays Wavefront Geometric interpretation Steering vector Array beam pattern Beam pattern in a plane Linear arrays Beam pattern symmetry for linear arrays Fourier transform interpretation Continuous Fourier transform approximation Sparse arrays Sparse arrays on a regular lattice Irregular random sparse arrays Polarization-diverse arrays Polarization formulation 196 Problems Angle-of-arrival estimation Maximum-likelihood angle estimation with known reference Maximum-likelihood angle estimation with unknown signal Beamscan Minimum-variance distortionless response MuSiC Example comparison of spatial energy estimators Local angle-estimation performance bounds Cramer Rao bound of angle estimation Cramer Rao bound: signal in the mean Cramer Rao bound: random signal Threshold estimation Types of transmitted signals Known reference signal test statistic Independent Rician random variables 221

11 Contents xi Correlated Rician random variables Unknown complex Gaussian signal Vector sensor 235 Problems MIMO channel Flat-fading channel Interference Maximizing entropy Flat-fading MIMO capacity Channel-state information at the transmitter Informed-transmitter (IT) capacity Uninformed-transmitter (UT) capacity Capacity ratio, c IT /c UT Frequency-selective channels Line-of-sight channel Stochastic channel models Spatially uncorrelated Gaussian channel model Spatially correlated Gaussian channel model Large channel matrix capacity Large-dimension Gaussian probability density Uninformed transmitter spectral efficiency bound Informed transmitter capacity Outage capacity SNR distributions Total power Fractional loss Channel estimation Cramer Rao bound Estimated versus average SNR Average SNR Estimated SNR MIMO capacity for estimated SNR in block fading Interpretation of various capacities Channel-state information at transmitter Reciprocity Channel estimation feedback 292 Problems Spatially adaptive receivers Adaptive spectral filtering Discrete Wiener filter Adaptive spatial processing Spatial matched filter Minimum-interference spatial beamforming 303

12 xii Contents MMSE spatial processing Maximum SINR SNR loss performance comparison Minimum-interference beamformer MMSE beamformer MIMO performance bounds of suboptimal adaptive receivers Receiver beamformer channel Iterative receivers Recursive least squares (RLS) Least mean squares (LMS) Multiple-antenna multiuser detector Maximum-likelihood demodulation Covariance matrix conditioning 337 Problems Dispersive and doubly dispersive channels Discretely sampled channel issues Noncommutative delay and Doppler operations Effect of frequency-selective fading Static frequency-selective channel model Frequency-selective channel compensation Eigenvalue distribution of space-time covariance matrix Space-time adaptive processing Orthogonal-frequency-division multiplexing Doubly dispersive channel model Doppler-domain representation Eigenvalue distribution of space-time-frequency covariance matrix Space-time-frequency adaptive processing Sparse space-time-frequency processing 362 Problems Space-time coding Rate diversity trade-off Probability of error formulation Outage probability formulation Block codes Alamouti s code Orthogonal space-time block codes Performance criteria for space-time codes Space-time trellis codes Trellis-coded modulation Space-time trellis coding Bit-interleaved coded modulation 381

13 Contents xiii Single-antenna bit-interleaved coded modulation Multiantenna bit-interleaved coded modulation Space-time turbo codes Direct modulation Universal codes Performance comparisons of space-time codes Computations versus performance 388 Problems Network Introduction Achievable rates of the 2 2 MIMO network Single-antenna Gaussian interference channel Achievable rates of the MIMO interference channel Outer bounds of the capacity region of the Gaussian MIMO interference channel Outer bounds to the capacity region of the single-antenna Gaussian interference channel Outer bounds to the capacity region of the Gaussian interference channel with multiple antennas The 2 2 cognitive MIMO network Non-cooperative primary link Cooperative primary link 412 Problems Cellular networks Point-to-point links and networks Multiple access and broadcast channels Linear receivers in cellular networks with Rayleigh fading and constant transmit powers Link lengths in cellular networks General network model Antenna-selection receiver Matched filter Linear minimum-mean-square-error receiver Laplacian of the interference Linear receivers in cellular networks with power control System model Optimality of parallelized transmissions with link CSI Asymptotic spectral efficiency of parallelized system Application to power-controlled systems without out-ofcell interference Monte Carlo simulations Matched-filter receiver in power-controlled cellular networks 448

14 xiv Contents Application to power-controlled systems with out-of-cell interference Summary 467 Problems Ad hoc networks Introduction Capacity scaling laws of ad hoc wireless networks Multiantenna links in ad hoc wireless networks Asymptotic spectral efficiency of ad hoc wireless networks with limited transmit channel-state information and minimum-mean-square-error (MMSE) receivers Spatially distributed network model Asymptotic spectral efficiency without transmit channel-state information Maximum-signal-to-leakage-plus-noise ratio receiver Linear receiver structures in spatially distributed networks Linear MMSE receivers in Poisson networks Laplacian of the interference in Poisson networks and matched-filter and antenna-selection receivers Interference alignment 487 Problems Medium-access-control protocols The need for medium-access control The ALOHA protocol Carrier-sense multiple access (CSMA) CSMA with collision avoidance (CSMA/CA) Non-space-division multiple-access protocols Space-division multiple-access (SDMA) protocols Introduction A simple SDMA protocol SPACE-MAC The reciprocity assumption Ward protocol Summary of some existing SDMA protocols 513 Problems Cognitive radios Cognitive radio channel Cooperative cognitive links Cognitive spectral scavenging Orthogonal-frequency-division multiple access Game-theoretical analysis 523

15 Contents xv 16.3 Legacy signal detection Known training sequence Single-antenna signal energy detection Multiple-antenna legacy signal detection Optimizing spectral efficiency to minimize network interference Optimal SISO spectral efficiency Optimal MIMO spectral efficiency 542 Problems Multiple-antenna acquisition and synchronization Flat-fading MIMO model Flat-fading MIMO delay-estimation bound Synchronization as hypothesis testing Motivations for test statistic approaches Test statistics for flat-fading channels Correlation MMSE beamformer Generalized-likelihood ratio test Spatial invariance Comparison of performance 557 Problems Practical issues Antennas Electrically small antennas Crossed polarimetric array Signal and noise model errors Noise figure Local oscillators Accuracy Phase noise Dynamic range Quantization Finite precision Analog nonlinearities Adaptive gain control Spurs Power consumption 568 References 569 Index 589

16

17 Preface In writing this text, we hope to achieve multiple goals. Firstly, we hope to develop a textbook that is useful as a reference for graduate or a supplement to advanced undergraduate classes investigating advanced wireless communications. These topics include adaptive antenna processing, multiple-input multipleoutput (MIMO) communications, and wireless networks. Throughout the text, there is a recurring theme of understanding and mitigating both internal and external interference. In addressing these areas of investigation, we explore concepts in information theory, estimation theory, signal processing, and implementation issues as are applicable. We attempt to provide a development covering these topics in a reasonably organized fashion. While not always possible, we attempt to be consistent in notation across the text. In addition, we provide problem sets that allow students to investigate these topics more deeply. Secondly, we attempt to organize the topics addressed so that this text will be useful as a reference. To the extent possible, each chapter will be reasonably self-contained, although some familiarity with the topic area is assumed. To aid the reader, reviews of many of the mathematical tools needed within the text are collected in Chapters 2 and 3. In addition, an overview of the basics of communications theory is provided in Chapters 4 and 5. Finally, in discussing these topics, we attempt to address a wide range of perspectives appropriate for the serious student of the area. Topics range from information theoretic bounds, to signal processing approaches, to practical implementation constraints. While there are many wonderful texts (and here we only list a subset) that address many of the topics of wireless communications [355, 280, 287, 314, 115, 324, 251, 255, 203], networks [100, 62], signal processing [275, 297, 238, 220, 204], array processing [294, 223, 205, 312, 248, 189], MIMO communications [247, 331, 160, 45, 22, 183, 84], information theory [68, 202, 212], estimation theory [312, 172, 297], and the serious researcher may wish to collect many of these texts, we hope that the particular collection and presentation of topics is uniquely useful to the research in advanced communications.

18 Acknowledgments I would like to thank and remember Professor David Staelin of Massachusetts Institute of Technology, whose interests and insights encouraged the authors to work together. I would like to thank my coauthor, who worked tirelessly with me to write this text. I would like to particularly thank Keith Forsythe of MIT Lincoln Laboratory, from whom I learned an immense amount over the years. A number of the concepts discussed in this text were developed by him or in collaboration with him. I will always be in debt for all that I learned from him. I would also like to thank (or blame) Jim Ward of MIT Lincoln Laboratory who encouraged me to write this text. Actually, I would like to thank everyone in the Advanced Sensor Techniques Group at MIT Lincoln Laboratory. I have learned something from every one of you. We thank the many individuals who have contributed comments and suggestions for the book: Pat Bidigare, Nick Chang, Glenn Fawcett, Jason Franz, Alan Fenn, Anatoly Goldin, Tim Hancock, Gary Hatke, Yichuan Hu, Scott Johnson, Josh Kantor, Nick Kaminski, Paul Kolodzy, Shawn Kraut, Raymond Louie, Adam Margetts, Matt McKay, Cory Myers, Peter Parker, Thomas Stahlbuhk, Vahid Tarokh, Gary Whipple, and Derek Young. In particular, we thank Bruce McGuffin and Ameya Agaskar who provided a significant number of comments. We would like to thank Dorothy Ryan for all her many helpful comments. To the folks at the Atomic Bean Cafe off of Harvard Square, thanks for all the espressos and for letting me spend many, many, many hours writing there. Finally, I would like to thank my family for their support. To my wife Nadya, and daughter Coco you may see more of me now. You can decide if that is good or bad. Dan Bliss Cambridge, MA

19 Acknowledgments xix I would like to thank and remember Professor David H. Staelin, formerly of the Massachusetts Institute of Technology for his inspiration, guidance and mentorship, and in particular for introducing me to my coauthor. I would like to thank my coauthor for his insight, mentorship and for being the driving force behind this book. I would also like to thank my former colleague at MIT, Danielle Hinton, in particular for insightful discussions on multiantenna protocols. I am grateful to my colleagues at Olin College including Brad Minch, Mark Somerville, and Vin Manno, for their encouragement and general discussions, both technical and non-technical. I would also like to thank my students and former students at Olin College, in particular Yifan Sun, Annie Martin, Rachel Nancollas, Katarina Miller, Jacob Miller, Jeff Hwang, Sean Shi, Elena Koukina, Yifei Feng, Rui Wang, Raghu Rangan, Tom Lamar, Avinash Uttamchandani, Ashley Lloyd, Junjie Zhu, and Chloe Egthebas for their direct and indirect contributions to this work, and in particular for helping me refine my presentation of some of the material that has made its way into the book. Finally, I would like to thank Alo, Antariksh, my parents, parents-in-law, siblings, and the rest of my family for their patience and tireless support. Siddhartan Govindasamy Natick, MA