Evaluation of HSDPA and LTE E-Book

Markus Rupp

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch
• Veröffentlichungsjahr: 2011

Seitenzahl: 638

Beschreibung

This book explains how the performance of modern cellular wireless networks can be evaluated by measurements and simulations With the roll-out of LTE, high data throughput is promised to be available to cellular users. In case you have ever wondered how high this throughput really is, this book is the right read for you: At first, it presents results from experimental research and simulations of the physical layer of HSDPA, WiMAX, and LTE. Next, it explains in detail how measurements on such systems need to be performed in order to achieve reproducible and repeatable results. The book further addresses how wireless links can be evaluated by means of standard-compliant link-level simulation. The major challenge in this context is their complexity when investigating complete wireless cellular networks. Consequently, it is shown how system-level simulators with a higher abstraction level can be designed such that their results still match link-level simulations. Exemplarily, the book finally presents optimizations of wireless systems over several cells. This book: * Explains how the performance of modern cellular wireless networks can be evaluated by measurements and simulations * Discusses the concept of testbeds, highlighting the challenges and expectations when building them * Explains measurement techniques, including the evaluation of the measurement quality by statistical inference techniques * Presents throughput results for HSDPA, WiMAX, and LTE * Demonstrates simulators at both, link- level and system-level * Provides system-level and link-level simulators (for WiMAX and LTE) on an accompanying website (href="https://www.nt.tuwien.ac.at/downloads/featured-downloads">https://www.nt.tuwien.ac.at/downloads/featured-downloads) This book is an insightful guide for researchers and engineers working in the field of mobile radio communication as well as network planning. Advanced students studying related courses will also find the book interesting.

Leseprobe

Table of Contents

Title Page

Copyright

About the Authors

About the Contributors

Preface

Acknowledgments

List of Abbreviations

Part One: Cellular Wireless Standards

Chapter 1: UMTS High-Speed Downlink Packet Access

1.1 Standardization and Current Deployment of HSDPA

1.2 HSDPA Principles

1.3 MIMO Enhancements of HSDPA

References

Chapter 2: UMTS Long-Term Evolution

2.1 LTE Overview

2.2 Network Architecture

2.3 LTE Physical Layer

2.4 MAC Layer

2.5 Physical, Transport, and Logical Channels

Referenes

Part Two: Testbeds for Measurements

Chapter 3: On Building Testbeds

3.1 Basic Idea

3.2 Transmitter

3.3 Receiver

3.4 Synchronization

3.5 Possible Pitfalls

3.6 Summary

References

Chapter 4: Quasi-Real-Time Testbedding

4.1 Basic Idea

4.2 Problem Formulation

4.3 Employing the Basic Idea

4.4 Data Collection

4.5 Evaluating and Summarizing the Data

4.6 Statistical Inference

4.7 Measurement Automation

4.8 Dealing with Feedback and Retransmissions

References

Part Three: Experimental Link-Level Evaluation

Chapter 5: HSDPA Performance Measurements

5.1 Mathematical Model of the Physical Layer

5.2 Receiver

5.3 Quantized Precoding

5.4 CQI and PCI Calculation

5.5 Achievable Mutual Information

5.6 Measurement Results

5.7 Summary

References

Chapter 6: HSDPA Antenna Selection Techniques

6.1 Existing Research

6.2 Receive Antenna Selection

6.3 An Exemplary Measurement and its Results

6.4 Summary

References

Chapter 7: HSDPA Antenna Spacing Measurements

7.1 Problem Formulation

7.2 Existing Research

7.3 Experimental Setup

7.4 Measurement Methodology

7.5 Measurement Results and Discussion

7.6 Different Transmit Power Levels and Scenarios

References

Chapter 8: Throughput Performance Comparisons

8.1 Introduction

8.2 Cellular Systems Investigated: WiMAX and HSDPA

8.3 Measurement Methodology and Setup

8.4 Measurement Results

8.5 Summary

References

Chapter 9: Frequency Synchronization in LTE

9.1 Mathematical Model

9.2 Carrier Frequency Offset Estimation in LTE

9.3 Performance Evaluation

References

Chapter 10: LTE Performance Evaluation

10.1 Mathematical Model of the Physical Layer

10.2 Receiver

10.3 Physical Layer Modeling

10.4 User Equipment Feedback Calculation

10.5 Practical Throughput Bounds

10.6 Simulation Results

References

Part Four: Simulators for Wireless Systems

Chapter 11: LTE Link- and System-Level Simulation

11.1 The Vienna LTE Link Level Simulator

11.2 The Vienna LTE System Level Simulator

11.3 Validation of the Simulators

11.4 Exemplary Results

References

Chapter 12: System-Level Modeling for MIMO-Enhanced HSDPA

12.1 Concept of System-Level Modeling

12.2 Computationally Efficient Link-Measurement Model

12.3 Link-Performance Model

References

Part Five: Simulation-Based Evaluation for Wireless Systems

Chapter 13: Optimization of MIMO-Enhanced HSDPA

13.1 Network Performance Prediction

13.2 RLC-Based Stream Number Decision

13.3 Content-Aware Scheduling

13.4 CPICH Power Optimization

References

Chapter 14: Optimal Multi-User MMSE Equalizer

14.1 System Model

14.2 Intra-Cell Interference Aware MMSE Equalization

14.3 The Cell Precoding State

14.4 Performance Evaluation

References

Chapter 15: LTE Advanced Versus LTE

15.1 IMT-Advanced and 3GPP Performance Targets

15.2 Radio Interface Enhancements

15.3 MIMO in LTE Advanced

15.4 Physical-Layer Throughput Simulation Results

References

Index

This edition first published 2012

© 2012 John Wiley & Sons Ltd

Registered office

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom

For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com.

The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.

Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.

MATLAB® is a trademark of The MathWorks, Inc. and is used with permission. The MathWorks does not warrant the accuracy of the text or exercises in this book. This book's use or discussion of MATLAB® software or related products does not constitute endorsement or sponsorship by The MathWorks of a particular pedagogical approach or particular use of the MATLAB® software.

Library of Congress Cataloging-in-Publication Data

Evaluation of HSDPA and LTE : from testbed measurements to system level performance / Sebastian Caban ... [et al.].

p. cm.

Includes bibliographical references and index.

ISBN 978-0-470-71192-7 (cloth)

1. Long-Term Evolution (Telecommunications) 2. Packet transmission (Data transmission) I. Caban, Sebastian.

TK5103.48325.E93 2012

621.382 – dc23

2011023059

A catalogue record for this book is available from the British Library.

ISBN: 9780470711927 (H/B)

ISBN: 9781119954699 (ePDF)

ISBN: 9781119954705 (oBook)

ISBN: 9781119960881 (ePub)

ISBN: 9781119960898 (mobi)

About the Authors

Sebastian Caban was born in Vienna, Austria, on March 13, 1980. After attending a technical high school and a mandatory service in the Austrian Armed Forces, he started studying electrical and communication Engineering at the Vienna University of Technology. He received his masters degree in 2005 and his Ph.D. degree in 2009 as “sub-auspiciis presidentis rei publicae.” In between, he studied a year at the University of Illinois at Urbana Champaign, USA. In 2005, Sebastian Caban also started to study economics at the Vienna University to relax from engineering. He completed a 5-year masters degree in business administration in 2010. Nonetheless, Sebastian Caban's work and research interest at the Vienna University of Technology focuses on developing measurement methodologies and building testbeds to quantify the actual performance of wireless cellular communication systems (UMTS, LTE) in realistic scenarios. Next to 12 national scholarships and prizes, Sebastian Caban was awarded the “Förderpreis 2006” of the German Vodafone Stiftung für Forschung for his research. E-mail: [email protected].

Christian Mehlführer was born in Vienna, Austria, in 1979. In 2004 he received his Dipl.-Ing. degree in electrical engineering from the Vienna University of Technology. Besides his diploma studies he worked part time at Siemens AG, where he performed integration tests of GSM carrier units. After finishing his diploma thesis on implementation and real-time testing of space–time block codes at the Institute of Communications and Radio-Frequency Engineering, Vienna University of Technology, for which he received the Vodafone Förderpreis 2006 (together with Sebastian Caban), he started his doctoral thesis at the same institute. In 2009, he finished his Ph.D. about measurement-based performance evaluation of WiMAX and HSDPA with summa cum laude. His research interests include experimental investigation of MIMO systems, UMTS HSDPA, WLAN (802.11), WiMAX (802.16), and the upcoming 3GPP LTE system. Currently, he is working as a consultant for Accenture. E-mail: [email protected].

Markus Rupp was born in 1963 in Völklingen, Germany. He received his Dipl.-Ing. degree in 1988 at the University of Saarbrücken, Germany and his Dr.-Ing. degree in 1993 at the Technische Universität Darmstadt, Germany, where he worked with Eberhardt Hänsler on designing new algorithms for acoustical and electrical echo compensation. From November 1993 until July 1995 he had a postdoctoral position at the University of Santa Barbara, California, with Sanjit Mitra, where he worked with Ali H. Sayed on a robustness description of adaptive filters with impact on neural networks and active noise control. From October 1995 until August 2001 he was a member of Technical Staff in the Wireless Technology Research Department of Bell-Labs at Crawford Hill, NJ, where he worked on various topics related to adaptive equalization and rapid implementation for IS-136, 802.11, and UMTS. Since October 2001 he has been a full professor for Digital Signal Processing in Mobile Communications at the Vienna University of Technology, where he served as Dean from 2005 to 2007. He was associate editor of IEEE Transactions on Signal Processing from 2002 to 2005, is currently associate editor of JASP EURASIP Journal of Advances in Signal Processing, and JES EURASIP Journal on Embedded Systems. He is a Senior Member of IEEE and elected AdCom member of EURASIP since 2004 and serving as president of EURASIP from 2009 to 2010. He has authored and co-authored more than 350 scientific papers, including 15 patents on adaptive filtering and wireless communications. E-mail: [email protected]

Martin Wrulich was born in 1980 in Klagenfurt, Carinthia. He received his Dipl.-Ing. degree from Vienna University of Technology in March 2006, which was awarded by the ‘Diplomarbeitspreis der Stadt Wien’. Furthermore, he received three times a scholarship from the faculty of electrical engineering and information technology for excellent academic performance. He started his doctoral studies in the area of system-level modeling and network optimization at the institute of communications and radio-frequency engineering in March 2006. In 2009, he received a Marshallplan scholarship to join the group of Prof. Paulraj at Stanford University, after which he finished his PhD ‘sub-auspiciis presidentis rei publicae’. He also acted as a TPC member for the IEEE broadband wireless access workshop in 2009. Currently he is working as a consultant for McKinsey&Company, Inc. E-mail: [email protected]

About the Contributors

Josep Colom Ikuno was born 1984 in Barcelona, Spain. He received his bachelor in Telecommunications Engineering from the Universitat Pompeu Fabra in Barcelona in 2005 and his Master's degree also in Telecom Engineering from the Universitat Politècnica de Catalunya in 2008, both with honors. Since beginning of 2008 he is with the Vienna University of Technology in the Mobile Communications Group, where he works towards his Ph.D thesis in the field of LTE link and system level simulation, modeling and optimization. His past and present research interests include video coding, OFDM systems, and accurate, low-complexity modeling of the physical layer for system level simulation of wireless networks, with emphasis on the integration of real operator data in the simulations and models. E-mail:[email protected]

José Antonio García-Naya was born in A Coruña, Spain, in June 1981. He obtained his master degree and his Ph.D. degree in Computer Engineering from the University of A Coruña in 2005 and 2010, respectively. Since 2005 he is with the Group of Electronic Technology and Communications (GTEC) at the Department of Electronics and Systems, University of A Coruña, Spain. From December 2007 to October 2008 he was a recipient of a Maria Barbeito post-degree grant from the Galician Government. Between 2008 and 2009 he was at the Institute of Communications and Radio-Frequency Engineering, Vienna University of Technology, Austria. He has been an Assistant Professor at the University of A Coruña since 2008. His research interests are in the field of digital communications, with special emphasis on prototyping of wireless communication equipment, hardware and software design of multiple antenna testbeds. E-mail: [email protected]

Qi Wang was born in Beijing, China, in November 1982. In 2001, she received the Bachelor degree in telecommunication engineering from Beijing University of Posts and Telecommunications. From August 2005 to October 2007, she studied in Linköping University in Linköping, Sweden, where she obtained the Master degree in computer science and engineering. Since November 2007, she works as university assistant at the Institute of Communications and Radio-Frequency Engineering of the Vienna University of Technology. Her research interests are in the field of wireless communications, in particular, synchronization aspects of OFDM systems. E-mail: [email protected].

Stefan Schwarz was born in 1984 in Neunkirchen, Austria. His interests in electronics and telecommunications were inspired while attending a technical high school. During this time, he was awarded the ‘GIT-Preis’ endowed by the ‘Austrian Electrotechnical Association’ for his thesis. He received a BSc degree in Electrical Engineering in 2007 and a Dipl.-Ing. degree in Telecommunications in 2009, both from Vienna University of Technology with highest distinction. Three times he received a scholarship from the faculty of electrical engineering and information technology for excellent academic performance, as well as the ‘Würdigungspreis’ from the Austrian federal department of science and research for the same reason. Since October 2009, he works as project assistant at the institute of telecommunications of the Vienna University of Technology, towards his Ph.D degree. His research interests are in wireless communications, currently focusing on link and network optimization and channel state information acquisition. E-mail: [email protected]

Preface

When you cannot measure it,

when you cannot express it in numbers,

your knowledge is of a meager

and unsatisfactory kind;

it may be the beginning of knowledge,

but you have scarcely, in your thoughts,

advanced to the stage of science.

William Thomson, Lord Kelvin (1894)

At the beginning of the 1990s, several research groups around the world were working on adaptive filters for hands-free telephones. At that time the first experimental DSP boards were available and well suited for audio experiments. While many research groups focused only on simulations, a few also implemented real-time experiments with such DSP boards. While such boards needed to be programmed in assembler, it was very tedious to make a program run and at the same time very time consuming, and many posed the question, “Why would experimental work be worth at all?” Why would it not be good enough to perform Matlab® simulations? In simulations one can be so much faster, have everything under control, and measurement precision is not an issue at all. However, experiments showed that simulations were not capturing the true nature of the problem. Some designs sounded artificial owing to their phase response degradation, some showed smaller Echo Return Loss Enhancement (ERLE) but simply sounded better. The “Quality of Experience” as we call it today was more important than the performance metric based on a Euclidian distance of ERLE. Also, rapid changes in the environment as they occur when the local speaker moves while speaking could never be simulated realistically. A simple linear combination from a starting impulse response to a final one was not capturing the motion. In conclusion, for this problem experimental work was crucial for its success, a success that we all experience today: hands-free telephones simply work!

At the beginning of the 1990s, similar problems occurred when designing the first cellular systems: the wireless channel and the multitude of elements in the transmission chain were largely mismodeled. At best we captured the qualitative behavior, never the true quantities. Even ray-tracing simulations, which are very demanding on the complexity, showed errors of 15 dB. Nevertheless, the efforts in experimental work at that time became marginal compared with simulation efforts. New transmission standards such as Universal Mobile Telecommunications System (UMTS) were entirely built on simulations. A 200 Mio investment for a new line of base stations was entirely built on simulations and not backed up by some physical demonstrations. By the time prototypes or demonstrators were built, the assembly line for the new-generation base station was already operational.

What is the reason that such a change in behavior has taken place? The first reason is certainly money, or, to be more precise, expected financial success. Whoever is first on the market makes the biggest profit. In wireless cellular systems the golden rule is “being six months early doubles profit, being six months late results in break-even.” Achieving results by simulations is a faster method than by experiments and, therefore, being favored. But there are other aspects that need to be mentioned. While in experiments you can never control all parameters, in simulations you can. Changing parameters as desired is possibly the greatest asset of simulations. Building a new setup for experiments can take a couple of weeks, if not months; changing two parameters in a simulation takes a minute. Next to flexibility there is also the aspect of time, as simulations are usually relatively fast. We use the word “relatively” for good reasons. Once the freedom exists to change parameters at free will, the requests for more and more simulations never stop. Also note that cellular systems become more and more complex and their complexity grows much faster than the available number crunching complexity on future computers (Shannon beats Moore). A third aspect is accuracy. Models become more and more detailed to achieve higher accuracy, but at the same time they become more complex with even more parameters. We can thus expect in the future that simulations will not give us as much benefit as they have given us in the past.

A strong argument for simulations has always been their flexibility. This has gone as far as some of our fellow researchers stating that only simulations can offer the potential to investigate a large variation of scenarios and thus allow for general statements, while experimental research is always limited to two or three setups that will never allow for generalization. We would like to argue strongly against such statement. First of all, models reflect the reality, but they are by far not reality. A good model is a simple model with very few parameters. As soon as the model requires several tens or even hundreds of parameters, it cannot be handled properly any more. A simulation including a variation of all parameters would not be feasible, and thus a conclusion about generality is not possible. This is exactly the situation we are in currently when dealing with modern cellular systems including multiple antennas at the transmitter and receiver end. Simple tapped delay line models (Pedestrian A, B) are not accurately reflecting reality; complex models (SCA, Winner) may include up to 100 parameters. Let us summarize our discussion with the following words:

Measurements reflect reality while simulations reflect the simulation environment.

We thus consider experimental work and validation to be of the utmost importance. Simply putting all faith in simulation results can only be good for the inhabitants of “SimCity,” but not for people living in the real world.

On the other hand, the drawbacks of experimental results are striking as well. Their little flexibility even over the design parameters of transmitter and receiver is often painful. In particular, when going for a prototype, the time for development might be as long as for the entire product and still not be as flexible as a simulation. A fair compromise seems to be our concept of testbeds. In testbeds the true physical channel is involved, as it is the part that is known the least. RF-front ends are typically replaced by relatively precise (but expensive) measurement equipment to allow for later evaluation of low-cost designs. The transmit signals are generated from a simulation environment such as Matlab in order to ensure identical data for simulation as well as experiments. Typically, receive signals are past processed in a non-real-time fashion. Such a testbed system thus provides almost the same flexibility as the simulation; only the wireless channel must be selected by the experimental setup. A further point is the automation of the measurement process utilizing XYΦ-positioning tables. Taking the manual aspects out and at the same time measuring several thousand locations (typically quarter wavelength spaced) makes the approach very valuable. Most time goes into the preparation of the experiment and the final postprocessing, allowing a larger measurement campaign, including postprocessing in 2–3 months, which is not so much longer than a modern simulation task of the same proportion. As computers become faster the simulation times will decrease, and so will the postprocessing task. As the received signals are fed back into the Matlab® chain, the results of the testbed measurement can be directly compared with simulation results, allowing the study of many effects in the more flexible simulation environment.

Our studies follow the idea that the transmission chain can be described by different kinds of losses. The first loss is named the “channel state information loss” or “feedback loss.” This is the difference between the channel capacity as described by Shannon and the mutual information of the channel. It can be interpreted as the loss a transmission must face when not knowing the channel at the transmitter end. As we show for Single-Input Single-Output (SISO) and cross-polarized antennas, such channel loss can be neglected. Only for four and more antenna systems does the channel loss take on considerable values. The second loss we encounter is called the “design loss.” If we take into account the part of the transmit power that we put in pilots and synchronization, or equivalently its corresponding part in bandwidth, then we can compute the so-called “achievable mutual information.” Owing to the design of the cellular system, even with perfect modulation and coding schemes, we will never exceed this bound. The difference between achievable mutual information and mutual information we thus call the design loss. Some cellular systems are well designed, such as High-Speed Downlink Packed Access (HSDPA), with small amounts in design loss, whereas Orthogonal Frequency-Division Multiplexing (OFDM) systems, such as Worldwide Inter-operability for Microwave Access (WiMAX) and Long-Term Evolution (LTE), exhibit large design losses. The final loss we encounter is the “implementation loss.” This is the difference between the measured throughput and the achievable mutual information. Such a loss is due to imprecise channel knowledge, to nonperfect modulation and coding schemes, to self- interference, as this is very strong in Wideband Code-Division Multiple Access (WCDMA) systems such as HSDPA, and all other known or unknown implementation aspects (oscillator phase noise, IQ imbalances, and many more). Our testbed approach allows the measurment of the first two losses, the channel loss and the design loss, perfectly. For the implementation loss we obtain a measure based on our realization including high-precision hardware. A low-cost product can expect to experience a higher implementation loss; thus, our measures provide a good idea of what potentially could be achieved. It is very educative to study the differences in design and implementation losses, as systems with higher design loss often compensate this with lower implementation loss. At the end, what counts is the sum of both. Let us end this discussion by another bon mot (Jan L. A. van de Snepscheut (1953–1994)).

In theory, there is no difference between theory and practice. But, in practice, there is.

How to read this book

Part One is a quick brush up on the cellular standards: High Speed Downlink Packet Access (HSDPA) and UMTS Long Term Evolution (LTE). Given the limited space in a book, the two chapters provide a compressed description of the physical layer with smaller add-ons of the upper layers. Basically everything required to understand the following four parts is provided for the reader. Experts can simply skip the chapters, whereas the engineer in the field working in a specific area may find a quick introduction into other standards. Students will find a comprehensible approach without becoming overloaded with too many details.

In Part Two we explain the concept of testbeds, as opposed to prototypes or demonstrators. We provide the challenges in building such testbeds and explain what to expect from them. Essentially, it turns out that simply combining subsystems does not make a testbed. The testbed explained here offers a very high flexibility, as essentially signals directly from Matlab® can be transmitted and similarly received signals can directly be fed back into the Matlab® simulator. This, in turn, offers a rigorous comparison of simulation and measurement results. Specifically, we focus on the measurement techniques, including evaluation of the measurement quality by statistical inference techniques as well as measurement automation, as only this step assures high-quality results in a moderate time frame. Finally, some first measurement examples are provided for indoor and outdoor measurements validating the measurement method.

Part Three reports on link-layer evaluation of cellular systems. We mostly focus on results obtained during several measurement campaigns in alpine and urban environments and show experimental results for HSDPA, and to some extent for WiMAX. In particular, we focus on questions of antenna distance when employing several antennas at the base station and antenna selection techniques, as our measurements show an entirely different picture than summation results of most publications, which are based on too simplistic channel assumptions. An entire chapter is related to the question of frequency synchronization, as this is a crucial issue typically not handled at all in the literature. As measurement results of LTE were not yet available in sufficient amounts at the time of writing this book, we relied on simulation results here. This part thus allows a direct comparison of the three technologies, not only in terms of throughput, but also from an information theoretical approach, by comparing the measured performance with mutual information and channel capacity. Moreover, we thoroughly investigated the reasons for the differences found; that is, we provide a detailed analysis of design and implementation losses. An important aspect also addressed here that is missing in most older literature is transmission with cross-polarized antenna systems. They not only provide higher capacity at a substantially smaller footprint, but they also offer substantial advantages for the signal processing part as the channel losses – that is, the differences between capacity and mutual information – become very small, offering to design and operate systems with very little feedback information. In theory, only the knowledge of the receiver SNR is required at the transmitter to achieve most of the capacity. The challenge is to design systems that achieve such quality.

In Parts Four and Five we take a different approach than in the previous parts; that is, the evaluation of cellular systems by simulation. In Part Four we explain how simulators are working at the link and system levels. Here, a particular focus is on system-level simulations, as such experiments would hardly be possible employing testbeds, since they require several base stations and a multitude of users. Particular focus is on the evaluation of such simulator tools. We demonstrate how this can be achieved by either comparing with cumulative density functions of measurements or by co-simulation of link and system-level parts. A simulator pair for link and system levels is introduced that was designed for co-simulation so that the link-level simulator can provide crucial information for the system level. Here, as in the previous chapters on measurements, we used modern statistical inference techniques, such as bootstrapping algorithms, to provide not only the average value (measured or simulated), but also always to complement our results with confidence intervals to make the statements much more meaningful.

Reproducibility has become an increasingly important issue in recent years. As systems become more and more complex, and thus complicated, it becomes more and more difficult to repeat results from others, and even reproducing our own results is often difficult after some time has passed. To facilitate reproducibility, therefore, we have launched open access1 “Vienna LTE Simulators” that can be downloaded, including many examples from our publications, and made to run on your own PC. After several ten thousands of downloads we also installed a forum to support the growing community. This gave us a lot of feedback and continuously smaller errors are found and corrected.

In Part Five we present evaluation results obtained from system-level simulators. On the example of HSDPA, we show how difficult and, at the same time, how complex such simulations are. With the available system-level simulator, several optimizations were performed: network performance prediction, Radio Link Control (RLC)-based stream number decision, content aware scheduling, Common Pilot CHannel (CPICH) power optimization. Finally, the last chapter deals with topics of LTE advanced standards, as they are being discussed at time of writing this book.

Sebastian Caban, Christian Mehlführer, Markus Rupp, Martin Wrulich

Notes

1 The open access, free of charge is granted for purely academic users. If a company or a project with a monetary flow is involved, a fee is required. The fee is used to further support these efforts.

Acknowledgments

This book is the outcome of many years of research and teaching in the field of signal processing and wireless communications.

We would like to thank many people from A1 (former Mobilkom and Telekom), as well as Kathrein for their steady support over a longer period of time: W. Wiedermann, W. Müllner, W. Karner, A. Ciaffone, T. Ergoth, W. Weiler, T. Baumgartner, J. Peterka, A. Kathrein, G. Schell, R. Gabriel, and J. Rumold.

A very special thank you goes to Christoph F. Mecklenbräuker, who supported many of our efforts via his Christian Doppler Laboratory “Wireless Technologies for Sustainable Mobility.” The HSDPA simulator was jointly developed with the ftw (Forschungszentrum Telekommunikation Wien). Many thanks for the support.

The work reported in this book required a lot of people that are usually not mentioned in publications, but without their administrative help we could not have done it: J. Auerböck, N. Hummer, W. Schüttengruber, B. Wistawel. Many thanks for your help.

Such a book would never have been possible without the constant support by the many helpful people from Wiley: T. Ruonamaa, A. Smart, S. Barclay, M. Cheok and including, G. Vasanth of Laserwords and P. Lewis. We cannot thank you enough.

List of Abbreviations

2G2nd Generation3GPP3rd Generation Partnership ProjectACKACKnowledgedALMMSEApproximate Linear Minimum Mean Square ErrorAMCAdaptive Modulation and CodingARQAutomatic Repeat reQuestAWGNAdditive White Gaussian NoiseBCaBias-Corrected and acceleratedBCCHBroadcast Control ChannelBCHBroadcast ChannelBICMBit-Interleaved Coded ModulationBERBit Error RatioBLEPBlock Error ProbabilityBLERBlock Error RatioCAContent AwareCAPEXCapital ExpenditureCBCode BlockCCChase CombiningCCCHCommon Control ChannelCDDCyclic Delay Diversitycdfcumulative density functionCDIChannel Direction IndicatorCDMACode-Division Multiple AccessCFOCarrier Frequency OffsetCLClosed LoopCLMIClosed-Loop Mutual InformationCLSMClosed-Loop Spatial MultiplexingCMCoded ModulationCoMPCoordinated Multi-PointCOSTEuropean COoperation in the field of Scientific and Technical researchCPCyclic PrefixC-PlaneControl PlaneCPICHCommon Pilot CHannelCPUCentral Processing UnitCQIChannel Quality IndicatorCRCCyclic Redundancy CheckCSIChannel State InformationCSI-RSCSI Reference SignalCTCConvolutional Turbo CodeCVQChannel Vector QuantizationCWCodewordDCCHDedicated Control ChannelDCHDedicated ChannelDFTDiscrete Fourier TransformDHCPDynamic Host Configuration ProtocolDiffServDifferentiate ServiceDLDownlinkDL-SCHDownlink Shared ChannelDMRSDemodulation Reference SignalsDoFsDegrees of FreedomDPCDirty Paper CodingDPCCHDedicated Physical Control CHannelDPDCHDedicated Physical Data CHannelDSDouble-StreamDSADynamic Sub-carrier AllocationDSTTD-SGRCDouble Space–Time Transmit Diversity with Sub-Group Rate ControlDTCHDedicated Traffic ChannelD-TxAADouble Transmit Antenna ArrayECREffective Code RateEDGEEnhanced Data Rates for Global system for mobile communications EvolutionEESMExponential Effective Signal to Interference and Noise Ratio MappingeNodeBEvolved base stationEPCEvolved Packet CoreESMEffective Signal to Interference and Noise Ratio MappingEUTRAEvolved Universal Terrestrial Radio AccessE-UTRANEvolved Universal Terrestrial Radio Access NetworkEVehAExtended Vehicular AFBIFeedBack InformationFDDFrequency Division DuplexFECForward Error CorrectionFFOFractional Frequency OffsetFFTFast Fourier TransformFIFOFirst In, First OutFPGAField-Programmable Gate ArrayFTPFile Transfer ProtocolGGSNGateway-General packet radio service Support NodeGOFGoodness Of FitGOPGroup Of PicturesGPRSGeneral Packet Radio ServiceGPSGlobal Positioning SystemGSCMGeometry-based Stochastic Channel ModelGSMGlobal System for Mobile communicationsHARQHybrid Automatic Repeat reQuestHSDPAHigh-Speed Downlink Packet AccessHS-DPCCHHigh-Speed Dedicated Physical Control CHannelHS-DSCHHigh-Speed Downlink Shared CHannelHSPAHigh-Speed Packet AccessHS-PDSCHHigh-Speed Physical Downlink Shared CHannelHSSHome Subscriber ServerHS-SCCHHigh-Speed Shared Control CHannelHSUPAHigh-Speed Uplink Packet AccessICInterference CancelationICIInter-Carrier InterferenceIRIncremental RedundancyIDEIntegrated Development EnvironmentIEEEInstitute of Electrical and Electronics EngineersIFFTInverse Fast Fourier TransformIFOInteger Frequency OffsetIMSIP Multimedia SubsystemIMTInternational Mobile TelecommunicationsIPInternet ProtocolIRIncremental RedundancyISIInter-Symbol InterferenceISMIndustrial, Scientific and MedicalISOInternational Standard OrganizationITUInternational Telecommunication UnionLANLocal Area NetworkLDPCLow Density Parity CheckLEPLink Error PredictionLLRLog-Likelihood RatioLMMSELinear Minimum Mean Square ErrorLMMSE-MAPLinear Minimum Mean Square Error Maximum A PosterioriLOSLine-Of-SightLSLeast SquaresLTELong-Term EvolutionLTE-ALong-Term Evolution-AdvancedMACMedium Access ControlMAC-dMedium Access Control dedicatedMAC-hsMedium Access Control for High-Speed Downlink Packet AccessMAPMaximum A-PosteriorimaxCIMaximum Carrier-to-Interference RatioMBMacroBlockMBMSMulticast Broadcast Multimedia ServicesMCCHMulticast Control ChannelMCHMulticast ChannelMCSModulation and Coding SchemeMIMutual InformationMIESMMutual Information Effective Signal to Interference and Noise Ratio MappingMIMOMultiple-Input Multiple-OutputMISOMultiple-Input Single-OutputMLMaximum LikelihoodMMEMobility Management EntityMMSEMinimum Mean Square ErrorMoRSEMobile Radio Simulation EnvironmentMPEGMoving Picture Expert GroupMRCMaximum Ratio CombiningMSEMean Square ErrorMTCHMulticast Control ChannelMTUMaximum Transfer UnitMUMulti-UserMU-MIMOMulti-User MIMOMVUMinimum Variance UnbiasedNACKNon-ACKnowledgedNALNetwork Abstract LayerNASNon-Access StratumNBAPNodeB Application PartNDINew Data IndicatorN-SAWN Stop And WaitNSNNokia Siemens NetworksNodeBBase stationOFDMOrthogonal Frequency-Division MultiplexingOFDMAOrthogonal Frequency-Division Multiple AccessOLMIOpen-Loop Mutual InformationOLSMOpen-Loop Spatial MultiplexingOOPObject-Oriented ProgrammingOPEXOperational ExpenditureOSIOpen Systems InterconnectionPAPRPeak-to-Average Power RatioPARCPer-Antenna Rate ControlPBCHPhysical Broadcast ChannelPCPersonal ComputerPCCCParallel Concatenated Convolutional CodePCCHPaging Control ChannelPCCPCHPrimary Common Control Physical ChannelPCHPaging ChannelPCIPrecoding Control IndicatorPCIsockPeripheral Component InterconnectPCRFPolicy and Charging Rules FunctionPDCCHPhysical Downlink Control ChannelPDCPPacket Data Convergence ProtocolPDSCHPhysical Downlink Shared Channelpdfprobability density functionPDPPacket Data ProtocolPDUPacket Data UnitPDNPacket Data NetworkPedAPedestrian APedBPedestrian BPERPacket Error RatioP-GWPDN GatewayPHYPhysicalPFProportional FairPMCHPhysical Multicast ChannelPMIPrecoding Matrix IndicatorPNPseudo NoisePPSPulse Per SecondPSCHPrimary Synchronization SignalPSDPower Spectral DensityPSKPhase-Shift KeyingPSNRPeak Signal to Noise RatioPRACHPhysical Random Access ChannelPUCCHPhysical Uplink Control ChannelPUSCHPhysical Uplink Shared ChannelQAMQuadrature Amplitude ModulationQBICMQuantized BICMQCIFQuarter Common Intermediate FormatQoEQuality of ExperienceQoSQuality of ServiceQPPQuadratic Permutation PolynomialQPSKQuadrature Phase Shift KeyingQSBICMQuantized and Shifted BICMRACHRandom Access ChannelRAIDRedundant Array of Independent DisksRANRadio Access NetworkRASReceive Antenna SelectionRBResource BlockREResource Elementrvidxredundancy version indexREResource ElementRFRadio FrequencyRFOResidual Frequency OffsetRIRank IndicatorRLCRadio Link ControlRMSRoot Mean SquareRNRelay NodeRNCRadio Network ControllerROIRegion Of InterestRRRound RobinRRCRadio Resource ControlRRCfiltRoot Raised CosineRRERemote Radio EquipmentRRMRadio Resource ManagementRS-CCReed–Solomon Convolutional CodeRS232Recommended Standard 232RTPReal-time Transport ProtocolRVRedundancy VersionRXReceiverSAESystem Architecture EvolutionSBICMShifted BICMSC-FDMASingle-carrier FDMASCHSynchronization ChannelSCMSpatial Channel ModelSCMESpatial Channel Model (SCM) ExtensionSDMASpatial Division Multiple AccessSDUService Data UnitSERSymbol Error RatioSGSNServing-General packet radio service Support NodeS-GWServing GatewaySICSuccessive Interference CancelationSIDSize Index IdentifierSIMOSingle-Input Multiple-OutputSINRSignal to Interference and Noise RatioSISOSingle-Input Single-OutputSMSpatial MultiplexingSNRSignal to Noise RatioSQPSequential Quadratic ProgrammingSSSingle-StreamSSCHSecondary Synchronization SignalSSDSoft Sphere DecoderSTBCSpace–Time Block CodeSTMMSESpace–Time Minimum Mean Squared ErrorSTTDSpace–Time Transmit DiversitySUSingle-UserSU-MIMOSingle-User MIMOSVDSingular Value DecompositionTASTransmit Antenna SelectionTBTransport BlockTBSTransport Block SizeTDDTime Division DuplexTFCTransport Format CombinationTFCITransport Format Combination IndicatorTFTTraffic Flow TemplateTOSType Of ServiceTPCTransmit Power-ControlTrCHTransport CHannelTSNTransmission Sequence NumberTTITransmission Time IntervalTXTransmitterTxAATransmit Antenna ArrayTxDTransmit DiversityUDPUser Datagram ProtocolUEUser EquipmentULUplinkULAUniform Linear ArrayUL-SCHUplink Shared ChannelUMTSUniversal Mobile Telecommunications SystemU-PlaneUser PlaneUSBUniversal Serial BusUTRAUniversal mobile telecommunications system Terrestrial Radio AccessUTRANUniversal mobile telecommunications system Terrestrial Radio Access NetworkV-BLASTVertical Bell Laboratories Layered Space–TimeVCEGVideo Coding Expert GroupVCLVideo Coding LayerVehAVehicular AVHSICVery High Speed Integrated CircuitsVoIPVoice over IPWB-CLMIWideband Closed-Loop Mutual InformationWB-CLMI-LRWideband Closed-Loop Mutual Information with Linear ReceiverWCDMAWideband Code-Division Multiple AccessWiMAXWorldwide Inter-operability for Microwave AccessWLANWireless Local Area NetworkZFZero Forcing

Part One

Cellular Wireless Standards

Introduction

Since the introduction of the Global System for Mobile communications (GSM) in 1991, the ecosystem of equipment vendors, application providers, and service enablers has grown significantly. Moreover, the development and availability of wireless broadband techniques now allows for an even tighter integration of web-based services on mobile devices. In the context of wireless networks, the key parameters defining the application performance include the data rate and the network latency. Some applications require only low bit rates of a few tens of kilobits per second but demand very low delay, as in Voice over IP (VoIP) and online games [4]. On the other hand, the download time of large files is only defined by the maximum data rate, and latency does not play a big role in this application.

In terms of penetration, mobile wireless telephony surpassed fixed-line volumes in 2004, whereas broadband coverage is still lagging behind [1]. Current surveys forecast a maximum penetration of approximately 25 % for European markets [5]. In Indonesia actually, High-Speed Downlink Packet Access (HSDPA) broadband access surpassed fixed broadband access in 2008. A recent study by Ericsson [2] claims that there are already 5 billion subscribers worldwide (July 2010), with a daily growth rate of 2 million; for 2020, some 50 billion subscribers are predicted. HSDPA played a big, if not the most important, role in the success of mobile broadband services. The dramatic push of the typical data rates for most services, as well as the achievable peak data rate compared with the Universal Mobile Telecommunications System (UMTS), together with the lowered latency, drew a lot of attention from customers. In addition, the rapid decline in prices [3] and the low-cost hardware to connect conveniently to HSDPA networks are very attractive for end users. Nowadays, no or only little effort is required to adapt internet applications to the mobile environment.

Essentially, HSDPA for the first time is a broadband wireless access with seamless mobility and extensive coverage. Together with High-Speed Uplink Packet Access (HSUPA), HSDPA forms the so-called High-Speed Packet Access (HSPA), which can be deployed on top of the existing Wideband Code-Division Multiple Access (WCDMA) UMTS networks, either on the same carrier (technically, this requires the split of the spreading code resources in the cell) or – for high capacity and high bit rate – on another carrier, thus consequently denoting a pure HSDPA operation. The ever-increasing demand for HSDPA broadband wireless access, however, drives mobile network operators to allocate their spectrum resources in a progressive way towards HSPA usage.

In this first part of the book, a short introduction of two wireless cellular standards is presented: HSDPA in Chapter 1 and the downlink mode of UMTS Long-Term Evolution (LTE) in Chapter 2. Given the limited space, the two chapters provide a compressed description of the physical layer with smaller add-ons of the upper layers, providing the reader with sufficient background information to understand the rest of the book. Experts can simply skip the chapters, whereas the engineer in the field working in a related area may find a quick introduction into such standards. Students of wireless telecommunications will find an easy-to-follow approach without becoming overloaded by too many details.

In Chapter 1 we particularly focus on the means for radio resource management, scheduling, as well as Multiple-Input Multiple-Output (MIMO) techniques. Already in Single-Input Single-Output (SISO) HSDPA, a so-called Channel Quality Indicator (CQI) has been introduced to allow the selection of optimal rate and modulation schemes depending on the observed channel situation. A small amount of feedback is thus returned to the transmitter, adjusting the data rate on the channel link. The aspect of multiple antennas comes with the notion of a precoding matrix which extends the concept of feedback information with a so-called Precoding Control Indicator (PCI). The topic on scheduling includes the functions of the Medium Access Control (MAC), in particular its advanced version Medium Access Control for High-Speed Downlink Packet Access (MAC-hs), as well as Radio Resource Management (RRM), and also covers the aspect of Hybrid Automatic Repeat reQuest (HARQ) in the form of Incremental Redundancy (IR) as applied in the HSDPA standard.

Such principles are also employed in LTE in Chapter 2, focusing more on the differences from the first chapter. In contrast to HSDPA, sets of subcarriers are combined in the form of Resource Blocks (RBs) in LTE. Adaptive Modulation and Coding (AMC) is being employed by means of the CQI as before and also the now-called Precoding Matrix Indicator (PMI) which takes on the role of the PCI parameter. But additionally, LTE now offers a Rank Indicator (RI) to select different modes of operation. The most common modes of operation, namely transmit diversity, Open-Loop Spatial Multiplexing (OLSM), and its counterpart Closed-Loop Spatial Multiplexing (CLSM), are explained.

References

[1] Commission of the European Communities (2008) ‘Progress report on the single European electronic communications market (14th report),’ Technical Report SEC (2009) 376, Commission of the European Communities. Available from http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2009:0140:FIN:EN:PDF.

[2] Ericsson (2010) ‘Mobile subscriptions hit 5 billion mark,’ Technical report, Ericsson. Available from http://www.ericsson.com/thecompany/press/releases/2010/07/1430616.

[3] Grinschgl, A. and Serentschy, G. (2007) ‘Kommunikationsbericht 2007,’ Technical report, Rundfunk und Telekom Regulierungs-GmbH. Available from http://www.rtr.at/de/komp/KBericht2007/K-Bericht_2007.pdf.

[4] Svoboda, P., Karner, W., and Rupp, M. (2007) ‘Traffic analysis and modeling for World of Warcraft,’ in Proceedings of IEEE International Conference on Communications (ICC), pp. 1612–1617. Available from http://publik.tuwien.ac.at/files/pub-et_12119.pdf.

[5] Wittig, H. (2008) ‘The power of mobile broadband – implications for European telcos and equipment vendors,’ Technical report, J.P. Morgan Securities Ltd.