Urban co-creation by Nexthamburg: the "Bremen on the move" (Bremen bewegen) project
Advanced Studies Mobile Research Center Bremen - …978-3-658-00808-6/1.pdf · Advanced Studies...
-
Upload
truongkien -
Category
Documents
-
view
213 -
download
1
Transcript of Advanced Studies Mobile Research Center Bremen - …978-3-658-00808-6/1.pdf · Advanced Studies...
Das Mobile Research Center Bremen (MRC) im Technologie-Zentrum Informatik
erprobt in enger Zusammenarbeit mit der Wirtschaft mobile Informatik-, Infor-mations- und Kommunikationstechnologien. Als Forschungs- und Transferinsti-tut des Landes Bremen vernetzt und koordiniert das MRC hochschulübergreifend eine Vielzahl von interdisziplinären Arbeitsgruppen, die sich mit der Entwicklung und Anwendung mobiler Lösungen beschäft igen. Die Reihe „Advanced Studies“ präsentiert ausgewählte hervorragende Arbeitsergebnisse aus der Forschungstätig-keit der Mitglieder des MRC.
In close collaboration with the industry, the Mobile Research Center Bremen (MRC), a division of the Center for Computing and Communication Technologies (TZI) of the University of Bremen, investigates, develops and tests mobile comput-ing, information and communication technologies. Th is research cluster of the state of Bremen links and coordinates interdisciplinary research teams from diff erent universities and institutions, which are concerned with the development and ap-plication of mobile solutions. Th e series “Advanced Studies” presents a selection of outstanding results of MRC’s research projects.
Edited byProf. Dr. Otthein HerzogProf. Dr. Carmelita GörgProf. Dr. Michael LawoMobile Research Center, Bremen, Germany
und Informationstechnik (TZI) der Universität Bremen erforscht, entwickelt und
Yasir ZakiBremen, Germany
Dissertation University of Bremen, 2012
ISBN 978-3-658-00807-9 ISBN 978-3-658-00808-6 (eBook) DOI 10.1007/978-3-658-00808-6
Th e Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografi e;detailed bibliographic data are available in the Internet at http://dnb.d-nb.de.
Library of Congress Control Number: 2012953090
Springer Vieweg© Springer Fachmedien Wiesbaden 2013 Th is work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifi cally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfi lms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, compu-ter soft ware, or by similar or dissimilar methodology now known or hereaft er developed. Exempted from this legal reservation are brief excerpts in connection with reviews or scho-larly analysis or material supplied specifi cally for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtai-ned from Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution under the respective Copyright Law. Th e use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specifi c statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. Th e publisher makes no warranty, express or implied, with respect to the material contained herein.
Printed on acid-free paper
Springer Vieweg is a brand of Springer DE. Springer DE is part of Springer Science+Business Media.www.springer-vieweg.de
Printed with friendly support of MRC Mobile Research Center, University of Bremen
Acknowledgments
It would not have been possible to finish this work and write the thesis withoutthe help and support of all the kind people around me, to only some of whom it ispossible to give particular mention here. I would like to thank my beloved wife Ta-mara Hamed, for her unconditional support and great patience. I will forever oweher a debt of gratitude for always believing in me, even in the times when I didn’t.I would also like to thank my parents, who have given me their un-equivalent sup-port, for which my mere expression of thanks likewise does not suffice. And to mylife joy, my kids, Tanya and Yezin.
This thesis would not have seen the light without the help and support of mysupervisor, Prof. Dr. Carmelita Görg. Her wide knowledge, encouragement andpersonal guidance have been of great value to me. I would also like to express mygratitude to Prof. Dr.-Ing. Andreas Timm-Giel, for all his valuable input.
The good advice, support and friendship of my colleague, Dr. Thushara Weera-wardane has been invaluable on both academic and personal level, for which I amextremely grateful. During this work I have collaborated with many colleagues forwhom I have great respect, in particular I wish to extend my warmest thanks toDr. Koojana Kuladinithi and Asanga Udugama for their professional and personalsupport. I would also like to express my gratitude to Dr. Hadeer Hamed and Mr.Khalis Mahmoud Khalis for their efforts in revising the thesis.
I would like to acknowledge all of my friends and colleagues within the Com-Nets department of the University of Bremen. Dr. Xi Li for her help in proofreading the thesis, Liang Zhao for being a great colleague in all the projects weworked together. Lots of gratitude to Dr. Andreas Könsgen, Markus Becker, UmarToseef, Muhammed Mutakin Siddique, Aman Singh, Dr. Bernd-Ludwig Wenning,Martina Kamman and Karl-Heinz Volk for their support. In addition, I would alsolike to thank my students Nikola Zahariev and Safdar Nawaz Khan Marwat fortheir support and good work.
Finally, a special acknowledgment for the DAAD (German Academic ExchangeService), for giving me the chance to come to Germany to do my Master studies,which led eventually to the finishing of this doctoral thesis.
Yasir Zaki
Abstract
Providing QoS while optimizing the LTE network in a cost efficient manner isvery challenging. Thus, radio scheduling is one of the most important functionsin mobile broadband networks. The design of a mobile network radio schedulerholds several objectives that need to be satisfied, for example: the scheduler needsto maximize the radio performance by efficiently distributing the limited radio re-sources, since the operator’s revenue depends on it. In addition, the scheduler hasto guarantee the user’s demands in terms of their Quality of Service (QoS). Thus,the design of an effective scheduler is rather a complex task. In this thesis, the au-thor proposes the design of a radio scheduler that is optimized towards QoS guar-antees and system performance optimization. The proposed scheduler is called“Optimized Service Aware Scheduler” (OSA). The OSA scheduler is tested andanalyzed in several scenarios, and is compared against other known schedulers.
A novel wireless network virtualization framework is also proposed in this the-sis. The framework targets the concepts of wireless virtualization applied withinthe 3GPP Long Term Evolution (LTE) system. LTE represents one of the new mo-bile communication systems that is just entering the market. Therefore, LTE waschosen as a case study to demonstrate the proposed wireless virtualization frame-work. The framework is implemented in the LTE network simulator and analyzed,highlighting the many advantages and potential gain that the virtualization processcan achieve. Two potential gain scenarios that can result from using network virtu-alization in LTE systems are analyzed: Multiplexing gain coming from spectrumsharing, and multi-user diversity gain.
Several LTE radio analytical models, based on Continuous Time Markov Chainsare designed and developed in this thesis. These models target the modelingof three different time domain radio schedulers: Maximum Throughput (MaxT),Blind Equal Throughput (BET), and Optimized Service Aware Scheduler (OSA).The models are used to obtain faster results (i.e., in a very short time period in theorder of seconds to minutes), compared to the simulation results that can take con-siderably longer periods, such as hours or sometimes even days. The model resultsare also compared against the simulation results, and it is shown that it provides agood match. Thus, it can be used for fast radio dimensioning purposes.
X Abstract
Overall, the concepts, investigations, and the analytical models presented in thisthesis can help mobile network operators to optimize their radio network and pro-vide the necessary means to support services QoS differentiations and guarantees.In addition, the network virtualization concepts provides an excellent tool that canenable the operators to share their resources and reduce their cost, as well as pro-vide good chances for smaller operators to enter the market.
Contents
Abstract IX
List of Figures XV
List of Tables XIX
List of Abbreviations XXI
List of Symbols XXV
1 Introduction 1
2 Mobile Communication Systems 5
2.1 Global System for Mobile Communication (GSM) . . . . . . . . 62.2 Universal Mobile Telecommunication System (UMTS) . . . . . . 9
3 Long Term Evolution (LTE) 13
3.1 Motivation and Targets . . . . . . . . . . . . . . . . . . . . . . . 133.2 LTE Multiple Access Schemes . . . . . . . . . . . . . . . . . . . 14
3.2.1 OFDM . . . . . . . . . . . . . . . . . . . . . . . . . . . 143.2.2 OFDMA . . . . . . . . . . . . . . . . . . . . . . . . . . 153.2.3 SC-FDMA . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.3 LTE Network Architecture . . . . . . . . . . . . . . . . . . . . . 173.3.1 User Equipment (UE) . . . . . . . . . . . . . . . . . . . 183.3.2 Evolved UTRAN (E-UTRAN) . . . . . . . . . . . . . . . 183.3.3 Evolved Packet Core (EPC) . . . . . . . . . . . . . . . . 19
3.4 E-UTRAN Protocol Architecture . . . . . . . . . . . . . . . . . . 203.4.1 Radio Link Control (RLC) . . . . . . . . . . . . . . . . . 233.4.2 Medium Access Control (MAC) . . . . . . . . . . . . . . 24
3.4.2.1 Logical and Transport Channels . . . . . . . . . 243.4.2.2 HARQ . . . . . . . . . . . . . . . . . . . . . . 253.4.2.3 Scheduling . . . . . . . . . . . . . . . . . . . . 26
XII Contents
3.4.3 LTE Frame and Physical Resource Structure . . . . . . . 273.5 LTE Quality of Service Bearers . . . . . . . . . . . . . . . . . . . 283.6 Beyond LTE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.6.1 Wider Bandwidth for Transmission . . . . . . . . . . . . 313.6.2 Advanced MIMO Solutions . . . . . . . . . . . . . . . . 313.6.3 CoMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323.6.4 Relays and Repeaters . . . . . . . . . . . . . . . . . . . . 32
4 LTE Network Simulator 35
4.1 Simulation Environment . . . . . . . . . . . . . . . . . . . . . . 354.2 Simulation Framework . . . . . . . . . . . . . . . . . . . . . . . 364.3 Simulation Model . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.3.1 UE Node Model . . . . . . . . . . . . . . . . . . . . . . 384.3.2 eNodeB Node Model . . . . . . . . . . . . . . . . . . . . 394.3.3 Access Gateway Node Model . . . . . . . . . . . . . . . 404.3.4 Packet Data Network Gateway Node Model . . . . . . . . 404.3.5 Mobility and Channel Node . . . . . . . . . . . . . . . . 414.3.6 Global User Database . . . . . . . . . . . . . . . . . . . 444.3.7 Application Configuration . . . . . . . . . . . . . . . . . 444.3.8 Profile Configuration . . . . . . . . . . . . . . . . . . . . 45
4.4 Traffic Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454.4.1 Voice over IP Model (VoIP) . . . . . . . . . . . . . . . . 464.4.2 Web Browsing Model . . . . . . . . . . . . . . . . . . . 474.4.3 Video Streaming Model . . . . . . . . . . . . . . . . . . 484.4.4 File Transfer Model . . . . . . . . . . . . . . . . . . . . 49
4.5 Statistical Evaluation . . . . . . . . . . . . . . . . . . . . . . . . 494.5.1 Confidence Interval Estimation . . . . . . . . . . . . . . . 494.5.2 Independent Replications Method . . . . . . . . . . . . . 50
5 LTE Virtualization 53
5.1 Virtualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535.1.1 Server Virtualization . . . . . . . . . . . . . . . . . . . . 545.1.2 Network Virtualization . . . . . . . . . . . . . . . . . . . 56
5.2 4WARD Project . . . . . . . . . . . . . . . . . . . . . . . . . . . 575.2.1 4WARD Virtualization Paradigm . . . . . . . . . . . . . . 57
5.3 Wireless Virtualization in Mobile Communication . . . . . . . . . 595.3.1 Motivation behind Mobile Network Virtualization . . . . . 615.3.2 LTE Virtualization Framework . . . . . . . . . . . . . . . 61
5.3.2.1 Framework Architecture . . . . . . . . . . . . . 62
Contents XIII
5.3.2.2 LTE Hypervisor Algorithm . . . . . . . . . . . 645.3.2.3 Operator Bandwidth Estimation . . . . . . . . . 645.3.2.4 Contract-Based Framework . . . . . . . . . . . 65
5.4 LTE Virtualization Evaluation . . . . . . . . . . . . . . . . . . . 685.4.1 Multiplexing Gain-Based Analysis . . . . . . . . . . . . . 695.4.2 Multi-User Diversity Gain-Based Analysis . . . . . . . . 735.4.3 Contract-Based Framework Analysis . . . . . . . . . . . 76
5.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
6 LTE Radio Scheduler 85
6.1 LTE Dynamic Packet Scheduling . . . . . . . . . . . . . . . . . . 866.2 LTE MAC Schedulers State of the Art . . . . . . . . . . . . . . . 87
6.2.1 Classical Scheduling Algorithms . . . . . . . . . . . . . . 876.3 Downlink MAC Scheduler Design . . . . . . . . . . . . . . . . . 89
6.3.1 QCI Classification . . . . . . . . . . . . . . . . . . . . . 906.3.2 Time Domain Scheduler (TDS) . . . . . . . . . . . . . . 916.3.3 Frequency Domain Scheduler (FDS) . . . . . . . . . . . . 946.3.4 Link-to-System Mapping (L2S) . . . . . . . . . . . . . . 966.3.5 HARQ Modeling . . . . . . . . . . . . . . . . . . . . . . 98
6.4 Downlink MAC Scheduler Analysis . . . . . . . . . . . . . . . . 996.4.1 OSA vs. Classical Schedulers . . . . . . . . . . . . . . . 99
6.4.1.1 FTP Only Scenario . . . . . . . . . . . . . . . 1006.4.1.2 Mixed Traffic Scenario . . . . . . . . . . . . . 104
6.4.2 GBR Delay Exploitation . . . . . . . . . . . . . . . . . . 1086.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
7 Analytical Modeling of the LTE Radio Scheduler 113
7.1 General Analytical Model . . . . . . . . . . . . . . . . . . . . . . 1147.1.1 Performance Evaluation . . . . . . . . . . . . . . . . . . 1157.1.2 Generic Departure Rate . . . . . . . . . . . . . . . . . . . 117
7.2 LTE Downlink Scheduler Modeling . . . . . . . . . . . . . . . . 1207.2.1 Single Class Model . . . . . . . . . . . . . . . . . . . . . 1207.2.2 Single Class Model Results . . . . . . . . . . . . . . . . 124
7.2.2.1 Analysis 1 - MaxT . . . . . . . . . . . . . . . . 1257.2.2.2 Analysis 2 - BET . . . . . . . . . . . . . . . . 1297.2.2.3 Analysis 3 - Sensitivity Analysis . . . . . . . . 1327.2.2.4 Analysis 4 - OSA . . . . . . . . . . . . . . . . 1337.2.2.5 Analysis 5 - BET with Random Direction (RD) 134
XIV Contents
7.2.3 Two-Dimensional Model . . . . . . . . . . . . . . . . . . 1367.2.3.1 2D TDS Modeling . . . . . . . . . . . . . . . . 1387.2.3.2 2D Performance Analysis . . . . . . . . . . . . 139
7.2.4 Two-Dimensional Model Results . . . . . . . . . . . . . . 1417.2.4.1 Analysis - w-MaxT . . . . . . . . . . . . . . . 1427.2.4.2 Analysis - w-BET . . . . . . . . . . . . . . . . 1467.2.4.3 Analysis - OSA . . . . . . . . . . . . . . . . . 150
7.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
8 Conclusions and Outlook 153
A Appendix 159
A.1 Mobility Models . . . . . . . . . . . . . . . . . . . . . . . . . . 159A.2 3GPP Transport Block Size . . . . . . . . . . . . . . . . . . . . . 161A.3 Simulation Results Confidence Interval . . . . . . . . . . . . . . 162
Bibliography 165
List of Figures
2.1 3GPP releases overview [HT09] [Zah11] . . . . . . . . . . . . . . 62.2 GSM network architecture 1 . . . . . . . . . . . . . . . . . . . . 72.3 GSM system evolution . . . . . . . . . . . . . . . . . . . . . . . 92.4 UMTS network architecture 1 . . . . . . . . . . . . . . . . . . . . 10
3.1 LTE EPS network architecture . . . . . . . . . . . . . . . . . . . 133.2 OFDM signal in frequency and time domain [Hoa05] . . . . . . . 153.3 An example of channel dependent scheduling between two users
[ADF+09] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163.4 An example comparing SC-FDMA to OFDMA [Agi09] . . . . . . 173.5 LTE E-UTRAN architecture . . . . . . . . . . . . . . . . . . . . 193.6 LTE E-UTRAN user plane protocol stack . . . . . . . . . . . . . 203.7 Detailed LTE downlink protocol architecture [Dah07] . . . . . . . 223.8 Multiple parallel HARQ processes example [Dah07] . . . . . . . 263.9 LTE FDD frame structure [Anr09] . . . . . . . . . . . . . . . . . 273.10 Relationship between slot, symbol and resource blocks [Anr09] . . 283.11 SAE bearer model [HT09] . . . . . . . . . . . . . . . . . . . . . 293.12 LTE-advanced CoMP . . . . . . . . . . . . . . . . . . . . . . . . 323.13 LTE-advanced in-band relay and backhaul [Agi11] . . . . . . . . 33
4.1 OPNET modeler© hierarchical editors . . . . . . . . . . . . . . . 354.2 LTE reference model . . . . . . . . . . . . . . . . . . . . . . . . 364.3 LTE OPNET simulation model . . . . . . . . . . . . . . . . . . . 374.4 UE node model . . . . . . . . . . . . . . . . . . . . . . . . . . . 384.5 eNodeB node model . . . . . . . . . . . . . . . . . . . . . . . . 394.6 aGW node model . . . . . . . . . . . . . . . . . . . . . . . . . . 404.7 PDN-GW node model . . . . . . . . . . . . . . . . . . . . . . . . 414.8 An example result of the fast fading model . . . . . . . . . . . . . 434.9 Sample OPNET application configuration . . . . . . . . . . . . . 444.10 Sample OPNET profile configuration . . . . . . . . . . . . . . . . 454.11 VoIP traffic model . . . . . . . . . . . . . . . . . . . . . . . . . . 46
XVI List of Figures
4.12 VoIP MOS values [IT09] . . . . . . . . . . . . . . . . . . . . . . 474.13 Web traffic model . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.1 Full virtualization environment [Cha09] . . . . . . . . . . . . . . 555.2 Para virtualization environment [Cha09] . . . . . . . . . . . . . . 555.3 OS virtualization environment [Cha09] . . . . . . . . . . . . . . . 565.4 Network virtualization proposed business model . . . . . . . . . . 585.5 Multiple access schemes . . . . . . . . . . . . . . . . . . . . . . 605.6 LTE virtualization framework architecture . . . . . . . . . . . . . 635.7 The general hypervisor algorithm framework . . . . . . . . . . . 675.8 LTE virtualization simulation model in OPNET . . . . . . . . . . 695.9 Virtual operators allocated bandwidth over simulation time . . . . 715.10 Virtual operator 1 VoIP air interface throughput . . . . . . . . . . 725.11 Virtual operator 1 VoIP application end-to-end delay . . . . . . . 725.12 Virtual operator 1 video application end-to-end delay . . . . . . . 735.13 Virtual operator 1 cell throughput with and without virtualization . 755.14 Virtual operator 1 cell throughput gain due to virtualization . . . . 765.15 Virtual operator allocated bandwidth/PRBs . . . . . . . . . . . . 785.16 Virtual operator 1 video downlink application end-to-end delay . . 795.17 Virtual operator 2 video downlink application end-to-end delay . . 805.18 Virtual operator 2 downlink allocated bandwidth . . . . . . . . . . 805.19 Virtual operator 3 video downlink application end-to-end delay . . 815.20 Virtual operator 3 average FTP download time . . . . . . . . . . . 825.21 Virtual operator 3 average number of FTP Files downloaded . . . 825.22 Virtual operator 4 VoIP downlink application end-to-end delay . . 83
6.1 General packet scheduling framework [HT09] . . . . . . . . . . . 866.2 OSA general scheduler framework . . . . . . . . . . . . . . . . . 906.3 FDS general flow chart . . . . . . . . . . . . . . . . . . . . . . . 956.4 Reference BLER versus SINR AWGN curves . . . . . . . . . . . 966.5 Average user FTP download time . . . . . . . . . . . . . . . . . . 1016.6 Unfairness between users FTP download time . . . . . . . . . . . 1026.7 Average cell throughput comparison . . . . . . . . . . . . . . . . 1036.8 40 UE scenario - scheduler comparison . . . . . . . . . . . . . . 1046.9 Application delay performance comparison between schedulers . . 1066.10 Fairness and cell throughput comparison between schedulers . . . 1076.11 Average cell throughput . . . . . . . . . . . . . . . . . . . . . . . 1096.12 Average VoIP application end-to-end delay . . . . . . . . . . . . 1106.13 Average VoIP MOS value . . . . . . . . . . . . . . . . . . . . . . 110
List of Figures XVII
6.14 non-GBR services performance comparison spider chart . . . . . 111
7.1 General Continuous Time Markov Chain (CTMC) . . . . . . . . . 1147.2 MaxT scheduling of users (example with 7 MCSs) . . . . . . . . 1227.3 BET scheduling of users (example with 7 MCSs) . . . . . . . . . 1227.4 MCSs static probability obtained from simulations . . . . . . . . 1237.5 OSA scheduling of users (example with 7 MCSs) . . . . . . . . . 1247.6 MaxT 10UEs scenario - average FTP download time . . . . . . . 1267.7 MaxT 20UEs scenario - average FTP download time . . . . . . . 1277.8 MaxT Markov chain state probability . . . . . . . . . . . . . . . 1287.9 BET 10UEs scenario - average FTP download time . . . . . . . . 1297.10 BET 20UEs scenario - average FTP download time . . . . . . . . 1307.11 BET Markov chain state probability . . . . . . . . . . . . . . . . 1317.12 MaxT sensitivity analysis results . . . . . . . . . . . . . . . . . . 1327.13 OSA 10UEs scenario - average FTP download time . . . . . . . . 1337.14 OSA Markov chain state probability . . . . . . . . . . . . . . . . 1347.15 BET RD 10UEs scenario - average FTP download time . . . . . . 1357.16 Two-dimensional Markov chain . . . . . . . . . . . . . . . . . . 1367.17 2D Markov chain represented in a single chain . . . . . . . . . . . 1397.18 Example Q mapping with N1=2 and N2=4 . . . . . . . . . . . . . 1397.19 Analysis1 w-MaxT average FTP download time . . . . . . . . . . 1427.20 Analysis2 w-MaxT average FTP download time . . . . . . . . . . 1437.21 Analysis1 w-MaxT Markov chain state probability . . . . . . . . 1447.22 Analysis2 w-MaxT Markov chain state probability . . . . . . . . 1457.23 Analysis3 w-BET average FTP download time . . . . . . . . . . . 1467.24 Analysis3 w-BET Markov chain state probability . . . . . . . . . 1477.25 Analysis4 w-BET average FTP download time . . . . . . . . . . . 1487.26 Analysis4 w-BET Markov chain state probability . . . . . . . . . 1497.27 Analysis5 OSA average FTP download time . . . . . . . . . . . . 150
A.1 Random Way Point (RWP) mobility model . . . . . . . . . . . . 159A.2 Random Direction (RD) mobility model . . . . . . . . . . . . . . 160
List of Tables
3.1 LTE MAC logical channels [36.11b] . . . . . . . . . . . . . . . . 253.2 LTE MAC transport channels [36.11b] . . . . . . . . . . . . . . . 253.3 LTE standardized QCIs and their parameters [SBT09] . . . . . . . 303.4 System performance comparison [Nak09] . . . . . . . . . . . . . 31
4.1 VoIP traffic model parameters [Li10] . . . . . . . . . . . . . . . . 464.2 Web browsing traffic model parameters [Wee11] . . . . . . . . . . 48
5.1 Scenario I simulation configurations . . . . . . . . . . . . . . . . 705.2 Scenario II simulation configurations . . . . . . . . . . . . . . . . 745.3 Scenario III simulation configurations . . . . . . . . . . . . . . . 77
6.1 DSCP/QCI to MAC-QoS-Class mapping example . . . . . . . . . 916.2 An example of QoS weight values for different non-GBR services 936.3 β values for each MCS [KSW+08][LV08a][Val06] . . . . . . . . 986.4 BLER and HARQ transmissions . . . . . . . . . . . . . . . . . . 996.5 Simulation configurations . . . . . . . . . . . . . . . . . . . . . . 1006.6 Simulation configurations . . . . . . . . . . . . . . . . . . . . . . 1056.7 Simulation configurations . . . . . . . . . . . . . . . . . . . . . . 108
7.1 n=2, K=2 combinations example . . . . . . . . . . . . . . . . . . 1197.2 Single class validation general parameters . . . . . . . . . . . . . 1247.3 Single class validation scenarios . . . . . . . . . . . . . . . . . . 1257.4 2-D model validation general parameters . . . . . . . . . . . . . . 1417.5 2-D model validation scenarios . . . . . . . . . . . . . . . . . . . 141
A.1 3GPP transport block size table (subset) . . . . . . . . . . . . . . 161A.2 Analysis 1 - MaxT 10 UEs simulation results confidence interval . 162A.3 Analysis 1 - MaxT 20 UEs simulation results confidence interval . 162A.4 Analysis 2 - BET 10 UEs simulation results confidence interval . . 162A.5 Analysis 2 - BET 20 UEs simulation results confidence interval . . 162
XX List of Tables
A.6 Analysis 3 - BET sensitivity analysis simulation results confidenceinterval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
A.7 Analysis 4 - OSA simulation results confidence interval . . . . . . 163A.8 Analysis 5 - BET RD simulation results confidence interval . . . . 163A.9 Analysis 1 - w-MaxT simulation results confidence interval . . . . 163A.10 Analysis 2 - w-MaxT simulation results confidence interval . . . . 163A.11 Analysis 3 - w-BET simulation results confidence interval . . . . 163A.12 Analysis 4 - w-BET simulation results confidence interval . . . . 163
List of Abbreviations
3GPP 3rd Generation PartnershipProject
AMC Adaptive Modulation andCoding
AODV Ad-hoc On-demand DistanceVector
ARP Allocation and RetentionPriority
ARQ Automatic Repeat Request
AUC Authentication Center
AWGN Additive White GaussianNoise
BET Blind Equal Throughput
BLER Block Error Rate
BSC Base Station Controller
BSS Base Station Subsystem
BTS Base Transceiver Station
CA Carrier Aggregation
CDMA Code Division MultipleAccess
CN Core Network
CoMP Coordinated Multi-Point
CQI Channel Quality Indicator
CTMC Continuous Time MarkovChain
DeNB Donor eNodeB
DL Downlink
DSCP Differentiated Services CodePoint
DVB Digital Video Broadcasting
eNodeB enhanced NodeB
E-UTRAN Evolved Universal TerrestrialRadio Access Network
EDGE Enhanced Data for GSMEvolution
EESM Exponential Effective SINRMapping
EFR Enhanced Full Rate
EMA Exponential Moving Average
EPC Evolved Packet Core
EPS Evolved Packet System
FDD Frequency Division Duplex
FDM Frequency DomainMultiplexing
FDMA Frequency Division MultipleAccess
FDS Frequency Domain Scheduler
FTP File Transfer Protocol
GBR Guaranteed Bit Rate
GMSK Gaussian Minimum ShiftKeying
GPRS General Packet Radio Service
GSM Global System for MobileCommunication
GTP Gateway Tunneling Protocol
XXII List of Abbreviations
HARQ Hybrid Automatic RepeatRequest
HLR Home Location Registry
HSDPA High Speed Downlink PacketAccess
HSPA High Speed Packet Access
HSS Home Subscriber Server
HSUPA High Speed Uplink PacketAccess
HTTP Hypertext Transfer Protocol
IMU International MobileTelecommunication
IP Internet Protocol
ISP Internet Service Provider
IT Information Technology
ITU InternationalTelecommunication Union
LTE Long Term Evolution
MAC Medium Access Channel
MaxT Maximum Throughput
MCS Modulation and CodingScheme
MIESM Mutual Information EffectiveSINR Mapping
MIMO Multi Input Multi Output
MME Mobility Management Entity
MOS Mean Opinion Score
MS Mobile Station
MSC Mobile Switching Center
non-GBR non-Guaranteed Bit Rate
OFDMA Orthogonal FrequencyDomain Multiple Access
OS Operating System
OSA Optimized Service Aware
PCRF Policy and Charging RulesFunction
PDA Personal Digital Assistant
PDCP Packet Data ConvergenceProtocol
PDN Packet Data Network
PDN-GW Packet Data Network Gateway
PDU Protocol Data Unit
PHY Physical Layer
PRB Physical Resource Block
PRBs Physical Resource Blocks
PSK Phase Shift Keying
PSTN Public Switched TelephoneNetwork
QCI QoS Class Identifier
QoS Quality of Service
RD Random Direction
RLC Radio Link Control
RN Relay Node
RNC Radio Network Controller
RRC Radio Resource Control
RWP Random Way Point
S-GW Serving Gateway
SAE System Architecture Evolution
SAN Storage Area Network
SC-FDMA Single Carrier FrequencyDomain Multiple Access
SDMA Space Division MultipleAccess
SDR Software Defined Radio
SIM Subscriber Identity Module
SINR Signal to Interference NoiseRatio
TBS Transport Block Size
TDS Time Domain Scheduler
List of Abbreviations XXIII
TDMA Time Division MultipleAccess
TE Terminal Equipment
TTI Transmission Time Interval
UE User Equipment
UL Uplink
UML User Mode Linux
UMTS Universal MobileTelecommunication System
USIM User Service Identity Module
UTRAN UMTS Terrestrial RadioAccess Network
VLR Visitor Location Registry
VM Virtual Machine
VMM Virtual Machine Monitor
VNet Virtual Network
VNOs Virtual Network Operators
VoIP Voice over Internet Protocol
WCDMA Wideband Code DivisionMultiple Access
WLAN Wireless Local Area Network
List of Symbols
Symbol Meaning
α smoothing factorβ MCS scaling factor
γk[t] normalized average channel condition of bearer kδ 2 varianceη actual number of users served per TTI
θmax maximum achieved throughput if all PRBs are used underperfect channel conditions
θk[t] instantaneous achieved throughput for bearer kθk[t] normalized average throughout of bearer k
λ arrival rate (file inter-arrival time)μ(n) generic departure rate of state nπ(n) state n steady state probability
πππ Markov chain steady state probability vectorτ smoothing factorψ maximum number of users served per TTI
BLEP([γk]) instantaneous Block Error Probability for channel state γk
BLEP([γe f f ]) instantaneous Block Error Probability for channel state γe f f
D mean number of departures by unit timeEtotal total BE PRB estimate over all BE operatorsE(N) average required PRBs at the Nth TTI
Fi operator i fairness factorHOLdelayk head-of-line packet delay for bearer k
K Number of MCSsMCSk kth modulation and coding scheme
n Number of active users per TTInk number of users in MCSk
N Number of users in the systemN0 thermal noise (dB)NF noise floor (dBm)
XXVI List of Symbols
Symbol Meaning
Pk MCSk static probabilityPL path loss (dB)Ptx eNodeB transmission power per PRB (dBm)
PBETk (t) BET scheduler time domain priority factor for bearer k
PGBRk (t) time domain GBR priority metric of bearer k
PMaxTk (t) MaxT scheduler time domain priority factor for user k
PnonGBRk (t) time domain non-GBR priority metric of bearer k
Pw−BETk (t) weighted BET scheduler time domain priority factor for user k
Pw−MaxTk (t) weighted MaxT scheduler time domain priority factor for user k
PRBsAlloci operator i allocated number of PRBsPRBsT T I(N) instantaneous PRB count at the Nth TTI
Q Markov chain infinitesimal generator matrixQ mean number of usersR distance between UE and eNodeB (km)
S(nδ ) slow fading at point nδ (dB)SINRe f f effective SINR mappingSINRk[t] instantaneous SINR value of bearer kSINRi, j Signal to Interference Noise Ratio on PRBi) for user j (dB)
SINRmax scaling factor (maximum achieved SINR)t(α/2,N−1) upper critical value of the t-distribution with N-1 degrees of freedom
to f f traffic model average OFF durationton average ON duration (file download time)
Tavg average download time of all usersTi per-user average download time
T BSk(η) number of bits that can be transmitted by a served UE using MCSk
T BS(n) state n average number of bits transmitted within a TTIT BS(n0, ...,nk) total bits sent by all served users under combination (n0, ...,nk)
UF% Unfairness factor (%)Vi i.i.d. normal random variable
WQoSj QoS weight of the jth MAC QoS classx sample mean
Xc de-correlation distance (m)Xk,i scheduler decision whether a UE is served or not (1 or 0)Xon traffic model average file size
Zα/2 upper α/2 critical value of the standard normal distribution