Handbook of Mobile Ad Hoc Networks for Mobility Models978-1-4419-6050-4/1.pdf · Handbook of Mobile...
Transcript of Handbook of Mobile Ad Hoc Networks for Mobility Models978-1-4419-6050-4/1.pdf · Handbook of Mobile...
Radhika Ranjan RoyUnited States Army ResearchDevelopment and Engineering Command
(RDECOM)Myer Center 2700Fort Monmouth, NJ 07703, [email protected]
ISBN 978-1-4419-6048-1 e-ISBN 978-1-4419-6050-4DOI 10.1007/978-1-4419-6050-4Springer New York Dordrecht Heidelberg London
Library of Congress Control Number: 2010934364
© Springer Science+Business Media, LLC 2011All rights reserved. This work may not be translated or copied in whole or in part without thewritten permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street,New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarlyanalysis. Use in connection with any form of information storage and retrieval, electronic adaptation,computer software, or by similar or dissimilar methodology now known or hereafter developed isforbidden.The use in this publication of trade names, trademarks, service marks, and similar terms, even if they arenot identified as such, is not to be taken as an expression of opinion as to whether or not they are subjectto proprietary rights.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)
To my Grandma for her causeless love, myparents Rakesh Chandra Roy and SneholotaRoy whose spiritual inspiration remainsvividly alive within all of us, my sistersGitaSree Roy, Anjali Roy, and Aparna Royand their spouses and my brother RaghunathRoy and his wife Nupur for their inspiration,my daughter Elora and my son-in-law Nick,my sons Ajanta and Debasri, and finally mybeloved wife Jharna for their love.
Preface
Throughout the course of my work in multihop mobile ad hoc networks (MANET)over the last several years, I reached the conclusion that mobility models and perfor-mance metrics need to be treated in detail in designing these networks that are theultimate frontier in wireless communications. A wide variety of mobility models canbe used by mobile nodes. Accurate representations of the characteristics of mobilenodes are key in understanding whether a given protocol used in the wireless com-munications network is useful in a particular type of ad hoc mobile scenario. Themobility performance metrics aim to capture the characteristics of different mobilitypatterns and can be used to analyze the performance of communications protocols.This book is an attempt to put together the theoretical aspects of the mobility modelsand metrics that are relevant to the mobile ad hoc network.
The mobility models are divided into seven different major groups based on theirbasic mobility characteristics: individual mobility, group mobility, autoregressivemobility, flocking mobility, virtual game-driven mobility, non-recurrent mobility,and time-variant community mobility. Many different variants of mobility modelsexist in each group and have been described in a chapter dedicated to each group.
The mobility performance metrics are grouped into seven major categories basedon the parameters that are being considered in each group: direct mobility met-rics, mobility measure metrics, link- and path-based metrics, network connectivitymetrics, quality-of-service metrics, energy performance metrics, and mobility pre-diction metrics. All mobility performance parameters for each group have beendescribed in each chapter.
The book has been organized into eight chapters: I. Introduction, II. IndividualMobility Models, III. Group Mobility Models, IV. Autoregressive Mobility Models,V. Flocking/Swarm Mobility Models, VI. Virtual Game-Driven Mobility Models,VII. Non-recurrent Mobility Models, and VIII. Social-Based Community MobilityModels. Readers familiar with mobile ad hoc networks will find interesting usingthe details of the mobility models and metrics in designing their wireless commu-nications networks where mobile nodes move from place to place with no fixedinfrastructures.
The book contains material from many remarkable published papers. I feel proudto mention the following authors as contributors whose material has been used in asubstantial way in the respective sections of the book as follows:
vii
viii Preface
• Section 3: T. Camp, J. Boleng, and V. Davies; M. McGuire; M. Piorkowski, N.Sarajanovoc-Djukic, and M. Grossglauser; C.A.V. Campos, D.C. Otero, and L.F. M. de Moraes
• Section 4: C. Bettstetter, H. Hartenstein and X. Perez-Costa; J. L. Boudec and M.Vojnovic; E. Hyyti, P. Lassila, and J. Virtamo
• Section 5: C. Bettstetter; T. Spyropoulos, A. Jindal, and K. Psounis; H. Liu, Y.Xu, and Q-A. Zeng
• Section 6: A. Jardosh, E. M. Belding-Royer, K. C. Almeroth and S. Suri;P. Venkateswaran, R. Ghosh, A. Das, S. K. Sanyal, and R. Nandi; H. M.Zimmermann, I. Gruber and C. Roman; M. Fiore, J. H¨arri, F. Filali, and C.Bonnet
• Section 7: B. Gloss, M. Scharf and D. Neubauer; E. Royer, P. M. Melliar-Smith,and L. Moser; Camp, J. Boleng, and V. Davies
• Section 8: T. Liu, P. Bahl, and I. Chlamtac; M. Zhao and W. Wang; K. Murrayand D. Pesch; D. Turgut, S. K. Das, and M. Chatterjee
• Section 9: T. Liu, P. Bahl, and I. Chlamtac; K. I. Smith, R. M. Eversion, and J. E.Fieldsend; M. Jamieson; R. Rao and G. Kesidis
• Section 10: T. Camp, J. Boleng, and V. Davies; B. Liang and Z. Haas• Section 11: M. Zhao and W. Wang• Section 12: Z. Haas; S. Cho and J. P. Hayes; T. Camp, J. Boleng, and V. Davies• Section 13: X. Li, B. Panja, and A. Zargari; H. Liu, Y. Xu, and Q-A. Zeng; W.
Shen, H. Liu, and Q-A. Zeng• Section 14: D. R. Basgeet, P. Dugenie, A. Munro, D. Kaleshi, and J. Irvine; S. C.
Nelson, A. F. Harris III, and R. Kravets• Section 15: X. Hong, T. Kwon, M. Gerla, D. Gu, and G. Pei; S. C. Nelson, A. F.
Harris III, and R. Kravets• Section 16: T. Cakar• Section 17: D. Helbing, I. J. Farkas, P. Molnar, and T. Vicsek; J. Gobel and A. E.
Krzesinski• Section 18: M. U. Ilyas and H. Radha• Section 19: F. Legendre, V. Borrel, M. D. de Amorim and S. Fdida• Section 20: R. Timo, K. Blackmore, L. Hanlen; S. Lim, C. Yu, and C. R. Das; S.
PalChaudhuri, J. L. Boudec and M. Vojnovic• Section 21: S. Bittner, W.-U. Raffel, and M. Scholz; J. Tian, J. Haehner, C.
Becker, I. Stepanov, and K. Rothermel• Section 22: X. Hong, M. Gerla, G. Pei, and C. Chiang; S. H. Manjula, C. N.
Abhilash, Shaila K., K. R. Venugopal, and L. M. Patnaik; T. Camp, J. Boleng,and V. Davies; S. S. Dalu, M. K. Naskar, and C. K. Sarkar
• Section 23: K. H. Wang and B. Li• Section 24: W. Chen and P. Chen• Section 25: K. Blakely and B. Lowekamp• Section 26: B. Zhou, K. Xu, and M. Gerla• Section 27: M. Rossi, L. Badia, N. Bui, and M. Zorzi• Section 28: V. Borrel, M. D. Amorim, and S. Fdida• Section 29: S. A. Williams and D. Huang• Section 30: J. Cano and P. Manzoni
Preface ix
• Section 31: Z. R. Zaidi, B. L. Mark, and R.K. Thomas• Section 32: Z. R. Zaidi, B. L. Mark, and R.K. Thomas• Section 33: R. Olfati-Saber• Section 34: D. S. Kim and S. K. Hwang• Section 35: F. Fitzek, L. Badia, M. Zorzi, G. Schulte, P. Seeling, and T.
Henderson; L. Petrak, O. Landsiedel, K. Wehrle• Section 36: S. Redon, Y. J. Kim, M. C. Lin, D. Manocha and J. Templeman; M.
C. Lin and D. Manocha; Y. Lu, H. Lin, Y. Gu and A. Helmy• Section 37: W. Hsu, T. Spyropoulos, K. Psounis, A. Helmy; C. Boldrini, M.
Conti, and A. Passarella• Section 38: J. Ghosh, S. Yoon, H. Ngo, and C. Qiao• Section 39: Y-X. Wang and F. S. Bao; H. Ochiai and H. Esaki; C. Gui; J. Leguay,
T. Timur Friedman, and V. Conan• Section 40: R. Sen, G. Hackmann, G. C. Roman, and C. Gill
Finally, I provide my heartfelt thanks to Susan Lagerstrom-Fife, Editor atSpringer, for helping me with the publication of my book in numerous ways includ-ing presenting material for outlining the publication proposal. My assistant editor,Jennifer Maurer, helped me in preparing the book manuscript and advised me withinfinite patience and good grace. I extend my thanks to Mr. Vignesh Kumar and histeam for in-depth proofing of the book.
Fort Monmouth, New Jersey Radhika Ranjan Roy
Contents
Part I Introduction
1 Mobile Ad Hoc Networks . . . . . . . . . . . . . . . . . . . . . . . 31.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.2 Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.3 Topology Control . . . . . . . . . . . . . . . . . . . . . . . 41.4 Medium Access . . . . . . . . . . . . . . . . . . . . . . . . 51.5 Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5.1 Unicast . . . . . . . . . . . . . . . . . . . . . . . 61.5.2 Broadcast . . . . . . . . . . . . . . . . . . . . . . 81.5.3 Multicast . . . . . . . . . . . . . . . . . . . . . . 91.5.4 Geocast . . . . . . . . . . . . . . . . . . . . . . . 11
1.6 Transport Protocol . . . . . . . . . . . . . . . . . . . . . . . 141.7 Quality of Service . . . . . . . . . . . . . . . . . . . . . . . 151.8 Energy Management . . . . . . . . . . . . . . . . . . . . . . 161.9 Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171.10 Mobile Peer-to-Peer Applications . . . . . . . . . . . . . . . 181.11 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191.12 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2 Mobility Model Characteristics . . . . . . . . . . . . . . . . . . . 232.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 232.2 Mobility Model Classifications . . . . . . . . . . . . . . . . 232.3 Formulation of Mobility Models . . . . . . . . . . . . . . . 252.4 Mobility Metrics . . . . . . . . . . . . . . . . . . . . . . . . 292.5 Impact of Mobility Models on MANET . . . . . . . . . . . 302.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
xi
xii Contents
Part II Individual Mobility Models
3 Random Walk Mobility . . . . . . . . . . . . . . . . . . . . . . . . 353.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 353.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383.3 Characteristics of Random Walk Mobility . . . . . . . . . . 393.4 Stationary Distribution of Random Walk Mobility . . . . . . 40
3.4.1 Steady-State Mobile Link Distribution . . . . . . . 413.4.2 Continuous-Time Steady-State Distribution
Approximation . . . . . . . . . . . . . . . . . . . 463.4.3 Simulation Results . . . . . . . . . . . . . . . . . 523.4.4 Summary . . . . . . . . . . . . . . . . . . . . . . 54
3.5 Limitations of Random Walk Mobility Model . . . . . . . . 543.6 Remedy of Limitations in Random Walk Mobility Model . . 553.7 Variations of Random Walk Mobility Model . . . . . . . . . 56
3.7.1 Markovian Random Walk Mobility . . . . . . . . . 573.7.2 Random Walk with Drift Mobility . . . . . . . . . 60
3.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 613.9 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4 Random Waypoint Mobility . . . . . . . . . . . . . . . . . . . . . 654.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 654.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674.3 Random Waypoint Stochastic Process . . . . . . . . . . . . . 674.4 Transition Length and Duration . . . . . . . . . . . . . . . . 69
4.4.1 Stochastic Process of Transition Lengths . . . . . . 704.4.2 Transition Length on 1D Line Segment . . . . . . . 714.4.3 Transition Length in Rectangular Area . . . . . . . 724.4.4 Transition Length in Circular Area . . . . . . . . . 754.4.5 Transition Time . . . . . . . . . . . . . . . . . . . 764.4.6 Time Between Two Direction Changes . . . . . . . 784.4.7 Spatial Node Distribution . . . . . . . . . . . . . . 784.4.8 Movement Direction . . . . . . . . . . . . . . . . 824.4.9 Boundary Changes . . . . . . . . . . . . . . . . . 864.4.10 Summary . . . . . . . . . . . . . . . . . . . . . . 89
4.5 Limitations of RWP Mobility Model . . . . . . . . . . . . . 904.6 Remedy of Limitations in RWP Mobility Model . . . . . . . 914.7 Variations of RWP Mobility Model . . . . . . . . . . . . . . 91
4.7.1 Notation for Generic Mobility Model . . . . . . . . 924.7.2 Generic Mobility Model . . . . . . . . . . . . . . 924.7.3 RWP on General Connected Domain . . . . . . . . 944.7.4 Restricted RWP . . . . . . . . . . . . . . . . . . . 964.7.5 RWP on Sphere . . . . . . . . . . . . . . . . . . . 984.7.6 RWP with Wrapping . . . . . . . . . . . . . . . . 994.7.7 RWP with Reflection . . . . . . . . . . . . . . . . 100
Contents xiii
4.7.8 Weighted Waypoint Mobility . . . . . . . . . . . . 1014.7.9 Summary . . . . . . . . . . . . . . . . . . . . . . 103
4.8 RWP Mobility with Arbitrary Waypoints . . . . . . . . . . . 1034.8.1 General Expressions for Traditional RWP . . . . . 1034.8.2 Spatial Node Distribution with Arbitrary Waypoints 1094.8.3 Connectivity in Mobile Ad Hoc Networks . . . . . 1164.8.4 Traffic Load in Dense Mobile Ad Hoc Networks . . 1184.8.5 Summary . . . . . . . . . . . . . . . . . . . . . . 120
4.9 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
5 Smooth Random Mobility . . . . . . . . . . . . . . . . . . . . . . 1255.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 1255.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1255.3 Characteristics of Smooth Random Mobility Model . . . . . 1265.4 Speed Control . . . . . . . . . . . . . . . . . . . . . . . . . 1275.5 Direction Control . . . . . . . . . . . . . . . . . . . . . . . 1295.6 Correlation Between Direction and Speed Change . . . . . . 131
5.6.1 Stop-Turn-and-Go Behavior . . . . . . . . . . . . 1315.6.2 Slowdown of Turning Nodes . . . . . . . . . . . . 132
5.7 Node Distribution and Border Behavior . . . . . . . . . . . . 1345.8 Encounter-Related Statistics for the Epoch-Based
Mobility Model . . . . . . . . . . . . . . . . . . . . . . . . 1355.8.1 Hitting and Meeting Times . . . . . . . . . . . . . 1355.8.2 Contact Duration and Inter-meeting Time . . . . . 1365.8.3 Assumptions . . . . . . . . . . . . . . . . . . . . . 137
5.9 Contact Time Statistics for the Epoch-Based MobilityModel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
5.10 Encounter Statistics for Smooth Random Mobility . . . . . . 1475.10.1 Hitting Time . . . . . . . . . . . . . . . . . . . . . 1475.10.2 Meeting Time . . . . . . . . . . . . . . . . . . . . 1485.10.3 Inter-meeting time . . . . . . . . . . . . . . . . . . 1505.10.4 Contact Duration . . . . . . . . . . . . . . . . . . 1515.10.5 Accuracy of the Analysis . . . . . . . . . . . . . . 152
5.11 Encounter-Related Statistics for the Epoch-Based RWP . . . 1535.11.1 Hitting and Meeting Times . . . . . . . . . . . . . 1535.11.2 Inter-meeting Time . . . . . . . . . . . . . . . . . 1545.11.3 Contact Duration . . . . . . . . . . . . . . . . . . 1545.11.4 Accuracy of the Analysis . . . . . . . . . . . . . . 157
5.12 Performance Analysis Using Encounter-RelatedStatistical Parameters . . . . . . . . . . . . . . . . . . . . . 1575.12.1 Simulation Environments . . . . . . . . . . . . . . 1595.12.2 Mobility-Assisted Routing Under No Contention . 1595.12.3 Mobility-Assisted Routing Under Contention . . . 161
5.13 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
xiv Contents
5.14 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
6 Geographic Constraint Mobility . . . . . . . . . . . . . . . . . . . 1676.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 1676.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1686.3 Vehicular Mobility . . . . . . . . . . . . . . . . . . . . . . . 169
6.3.1 Freeway Mobility . . . . . . . . . . . . . . . . . . 1696.3.2 Manhattan Grid Mobility . . . . . . . . . . . . . . 1736.3.3 Car-Following Mobility . . . . . . . . . . . . . . . 1746.3.4 Vehicular Mobility Model Analysis . . . . . . . . . 1776.3.5 Impact of Vehicular Mobility on MANET . . . . . 1816.3.6 Summary . . . . . . . . . . . . . . . . . . . . . . 184
6.4 Obstacle Mobility . . . . . . . . . . . . . . . . . . . . . . . 1856.4.1 Obstacle Construction . . . . . . . . . . . . . . . . 1856.4.2 Voronoi Tessellation and Pathways . . . . . . . . . 1866.4.3 Semi-definitive Node Movement . . . . . . . . . . 1886.4.4 Exponentially Distributed Destination Selection . . 1896.4.5 Attraction Point Movement . . . . . . . . . . . . . 1906.4.6 Simulations . . . . . . . . . . . . . . . . . . . . . 1906.4.7 Simulation Results . . . . . . . . . . . . . . . . . 1946.4.8 Summary . . . . . . . . . . . . . . . . . . . . . . 202
6.5 Community-Based Obstacle Mobility . . . . . . . . . . . . . 2036.5.1 COM Model Characteristics . . . . . . . . . . . . 2046.5.2 Mobility Controlling Criteria . . . . . . . . . . . . 2046.5.3 Pause-Time Criteria . . . . . . . . . . . . . . . . . 2086.5.4 Movement in Presence of Obstacles . . . . . . . . 2086.5.5 Simulation Results and Analysis . . . . . . . . . . 2096.5.6 Summary . . . . . . . . . . . . . . . . . . . . . . 212
6.6 Voronoi-Based Mobility . . . . . . . . . . . . . . . . . . . . 2126.6.1 Voronoi Environment Model . . . . . . . . . . . . 2136.6.2 Voronoi Mobility Model Characteristics . . . . . . 2176.6.3 Summary . . . . . . . . . . . . . . . . . . . . . . 219
6.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
7 Realistic Random Direction Mobility . . . . . . . . . . . . . . . . 2237.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 2237.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2247.3 Random Direction Mobility Model Characteristics . . . . . . 2257.4 Random Direction Model with Location-Dependent
Parameterization . . . . . . . . . . . . . . . . . . . . . . . . 2267.4.1 Impact of Location-Dependent Parameterization . . 2297.4.2 Automated Generation of Parameterizations . . . . 230
7.5 Modified Random Direction Mobility . . . . . . . . . . . . . 235
Contents xv
7.6 Comparison Between RDM, MRD and RWPMobility Models . . . . . . . . . . . . . . . . . . . . . . . . 2357.6.1 Simulation Environment . . . . . . . . . . . . . . 2357.6.2 Simulations . . . . . . . . . . . . . . . . . . . . . 2367.6.3 Results . . . . . . . . . . . . . . . . . . . . . . . . 2407.6.4 Summary . . . . . . . . . . . . . . . . . . . . . . 242
7.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
8 Deterministic Mobility . . . . . . . . . . . . . . . . . . . . . . . . 2458.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 2458.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2468.3 Mathematical Formulation of Deterministic Mobility . . . . . 2468.4 DMM with Constant Speed and Direction . . . . . . . . . . . 2478.5 DMM with Known Mobility Patterns . . . . . . . . . . . . . 248
8.5.1 Global Mobility Model . . . . . . . . . . . . . . . 2488.5.2 Global Prediction Algorithm Using GMM . . . . . 2508.5.3 Simulation and Results . . . . . . . . . . . . . . . 2528.5.4 Summary . . . . . . . . . . . . . . . . . . . . . . 252
8.6 Purposeful Deterministic Mobility . . . . . . . . . . . . . . 2528.6.1 Formulation for Purposeful Deterministic
Mobility Model . . . . . . . . . . . . . . . . . . . 2538.6.2 Summary . . . . . . . . . . . . . . . . . . . . . . 257
8.7 DMM with Attraction Points . . . . . . . . . . . . . . . . . 2578.7.1 Mobility Model . . . . . . . . . . . . . . . . . . . 2588.7.2 Simulation Study . . . . . . . . . . . . . . . . . . 2598.7.3 Summary . . . . . . . . . . . . . . . . . . . . . . 261
8.8 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
9 Partially Deterministic Mobility . . . . . . . . . . . . . . . . . . . 2659.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 2659.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2669.3 PDM with Known Direction of Movement . . . . . . . . . . 2689.4 PDM with Known Mobility Patterns . . . . . . . . . . . . . 268
9.4.1 Local Mobility Model . . . . . . . . . . . . . . . . 2699.4.2 Hierarchical Location Prediction Algorithm . . . . 2759.4.3 Simulation and Results . . . . . . . . . . . . . . . 2819.4.4 Prediction Performance . . . . . . . . . . . . . . . 2869.4.5 Systems Implementation . . . . . . . . . . . . . . 2889.4.6 Summary . . . . . . . . . . . . . . . . . . . . . . 291
9.5 Purposeful Partially Deterministic Mobility . . . . . . . . . . 2929.5.1 PPD Mobility Model . . . . . . . . . . . . . . . . 2939.5.2 Dual Tasking: Scanning and Relaying . . . . . . . 3019.5.3 Simulation Study . . . . . . . . . . . . . . . . . . 3019.5.4 Summary . . . . . . . . . . . . . . . . . . . . . . 306
xvi Contents
9.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
10 Random Gauss–Markov Mobility . . . . . . . . . . . . . . . . . . 31110.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 31110.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31210.3 RGM Mobility Model Description . . . . . . . . . . . . . . 313
10.3.1 One-Dimensional Case . . . . . . . . . . . . . . . 31310.3.2 Multi-dimensional Case . . . . . . . . . . . . . . . 31710.3.3 RGM and RWP Mobility Models . . . . . . . . . . 31810.3.4 RGM Mobility Tracking . . . . . . . . . . . . . . 31810.3.5 RGM Model Parameter Estimation . . . . . . . . . 319
10.4 RGM Mobility in Cellular Wireless Network . . . . . . . . . 32010.4.1 Predictive Location Updating and Selective
Paging Scheme . . . . . . . . . . . . . . . . . . . 32210.4.2 Cost Estimation for Mobility Management . . . . . 32310.4.3 One-Dimensional Cost Evaluation . . . . . . . . . 32410.4.4 Two-Dimensional Cost Evaluation . . . . . . . . . 32910.4.5 Two-Dimensional Cost Approximation . . . . . . . 33010.4.6 Numerical Results and Comparisons . . . . . . . . 33110.4.7 Network Parameter Optimization with Ideal
RGM Mobility . . . . . . . . . . . . . . . . . . . 33210.4.8 Comparison with Non-predictive
Distance-Based Scheme . . . . . . . . . . . . . . . 33510.4.9 Dynamic RGM Mobility Parameter Estimation . . 340
10.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34210.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
11 Semi-Markov Smooth Mobility . . . . . . . . . . . . . . . . . . . 34511.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 34511.2 Notations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34511.3 Semi-Markov Smooth Mobility Model Characteristics . . . . 347
11.3.1 Speedup Phase (α-Phase) . . . . . . . . . . . . . . 34711.3.2 Middle Smooth Phase (β-Phase) . . . . . . . . . . 34711.3.3 Slowdown Phase (γ -Phase) . . . . . . . . . . . . . 34911.3.4 Semi-Markov Process . . . . . . . . . . . . . . . . 349
11.4 Average Steady-State Speed . . . . . . . . . . . . . . . . . . 35011.5 Uniform Node Distribution . . . . . . . . . . . . . . . . . . 35111.6 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . 352
11.6.1 Simulation Setup . . . . . . . . . . . . . . . . . . 35211.6.2 Average Speed . . . . . . . . . . . . . . . . . . . . 35311.6.3 Spatial Node Distribution . . . . . . . . . . . . . . 35411.6.4 Comparison . . . . . . . . . . . . . . . . . . . . . 354
11.7 SMS Mobility Model Properties . . . . . . . . . . . . . . . . 35511.7.1 Smooth Movement . . . . . . . . . . . . . . . . . 355
Contents xvii
11.7.2 Stable Average Speed . . . . . . . . . . . . . . . . 35611.7.3 Uniform Spatial Node Distribution . . . . . . . . . 356
11.8 Applications of SMS Mobility Model . . . . . . . . . . . . . 35711.8.1 Routing Performance . . . . . . . . . . . . . . . . 35711.8.2 Network Connectivity . . . . . . . . . . . . . . . . 35811.8.3 Group Mobility . . . . . . . . . . . . . . . . . . . 36011.8.4 Geographic Constrained Networks . . . . . . . . . 361
11.9 Topology Dynamics Analysis . . . . . . . . . . . . . . . . . 36311.9.1 Relative Movement Trajectory Modeling . . . . . . 36311.9.2 Distance Transition Probability Matrix . . . . . . . 36411.9.3 Link Lifetime . . . . . . . . . . . . . . . . . . . . 36811.9.4 Link Change Rate . . . . . . . . . . . . . . . . . . 36911.9.5 Network Connectivity . . . . . . . . . . . . . . . . 372
11.10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37411.11 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
12 Boundless Simulation Area Mobility . . . . . . . . . . . . . . . . 37912.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 37912.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37912.3 BSA Mobility Model Description . . . . . . . . . . . . . . . 38112.4 BSA Mobility Model Evaluation . . . . . . . . . . . . . . . 38312.5 Uses of BSA in Mobile Networks . . . . . . . . . . . . . . . 384
12.5.1 Zone Routing Protocol . . . . . . . . . . . . . . . 38412.5.2 Connection Stability with Constant Velocity . . . . 389
12.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40112.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
13 Fluid-Flow Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . 40513.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 40513.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40513.3 Fluid-Flow Mobility Model Description . . . . . . . . . . . 408
13.3.1 Classical Fluid-Flow Mobility . . . . . . . . . . . 40813.3.2 Dynamic Fluid-Flow Mobility . . . . . . . . . . . 408
13.4 Applications of Fluid-Flow Mobility Model . . . . . . . . . 41013.4.1 Cellular Wireless Networks . . . . . . . . . . . . . 41013.4.2 Mobile Ad Hoc Networks . . . . . . . . . . . . . . 425
13.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43813.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
14 Gravity Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44314.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 44314.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44314.3 Simple Gravity Mobility Model . . . . . . . . . . . . . . . . 445
xviii Contents
14.4 Gravity Mobility Model Extensions . . . . . . . . . . . . . . 44614.4.1 Parameter Calibration . . . . . . . . . . . . . . . . 44714.4.2 Example Methodology . . . . . . . . . . . . . . . 447
14.5 Gravity-Based Composite Mobility Model . . . . . . . . . . 44914.5.1 Scalable Mobility Model . . . . . . . . . . . . . . 44914.5.2 Disaster Mobility Model . . . . . . . . . . . . . . 467
14.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482
15 Mobility Vector Model . . . . . . . . . . . . . . . . . . . . . . . . 48315.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 48315.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48315.3 Mobility Vector Model Description . . . . . . . . . . . . . . 48415.4 Mobility Vector Framework and Other Mobility Models . . . 485
15.4.1 Gravity Mobility . . . . . . . . . . . . . . . . . . 48515.4.2 Location-Dependent Mobility . . . . . . . . . . . . 48515.4.3 Targeting Mobility . . . . . . . . . . . . . . . . . 48515.4.4 Group Motion Mobility . . . . . . . . . . . . . . . 485
15.5 Calibration of Mobility Parameters . . . . . . . . . . . . . . 48615.5.1 Average Speed and Distance Traveled . . . . . . . 48615.5.2 Transmission Range and Link Change Rate . . . . 486
15.6 Impact on Network Performance Analysis . . . . . . . . . . 48915.6.1 Experimental Configuration . . . . . . . . . . . . . 49015.6.2 Results . . . . . . . . . . . . . . . . . . . . . . . . 490
15.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49215.8 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493
16 Correlated Diffusion Mobility . . . . . . . . . . . . . . . . . . . . 49516.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 49516.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49616.3 Correlated Diffusion Mobility Model Description . . . . . . 497
16.3.1 Correlation of Motion Components . . . . . . . . . 49716.3.2 Two-Dimensional-Correlated Random Walk . . . . 49816.3.3 Random Walk Statistics . . . . . . . . . . . . . . . 49916.3.4 Limiting Behavior of Random Walk . . . . . . . . 50016.3.5 Joint Normal Solution . . . . . . . . . . . . . . . . 50216.3.6 Statistical Description of Residence Time . . . . . 504
16.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52316.5 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 524References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525
17 Particle-Based Mobility . . . . . . . . . . . . . . . . . . . . . . . . 52717.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 52717.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527
Contents xix
17.3 Particle-Based Mobility Using Newtonian Mechanics . . . . 52917.4 Generalized Particle-Based Mobility and Concept
of Behavioral Forces . . . . . . . . . . . . . . . . . . . . . . 52917.5 Application of the Particle-Based Mobility Model
in Pedestrian Dynamics . . . . . . . . . . . . . . . . . . . . 53117.5.1 Simulation Results . . . . . . . . . . . . . . . . . 53317.5.2 Summary . . . . . . . . . . . . . . . . . . . . . . 543
17.6 Particle-Based Mobility Application in MANET . . . . . . . 54417.6.1 Particle-Based Mobility Using Newton’s
Gravitational Law . . . . . . . . . . . . . . . . . . 54517.6.2 Simulation Results . . . . . . . . . . . . . . . . . 54717.6.3 Summary . . . . . . . . . . . . . . . . . . . . . . 554
17.7 Particle-Based Mobility Using Quantum Mechanics . . . . . 55517.8 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 556
18 Hierarchical Influence Mobility . . . . . . . . . . . . . . . . . . . 55718.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 55718.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55818.3 Features of Hierarchical Influence Mobility Model . . . . . . 55918.4 Graph-Based HIMM . . . . . . . . . . . . . . . . . . . . . . 56018.5 Binary HIMM . . . . . . . . . . . . . . . . . . . . . . . . . 56218.6 Evil Rain HIMM . . . . . . . . . . . . . . . . . . . . . . . . 56218.7 Mobility Simulation Scenarios . . . . . . . . . . . . . . . . 563
18.7.1 Pedestrian Crossing . . . . . . . . . . . . . . . . . 56318.7.2 Intra-state Travel . . . . . . . . . . . . . . . . . . 566
18.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56818.9 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 569References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 569
19 Behavioral Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . 57119.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 57119.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57219.3 Behavioral Mobility Model Paradigm . . . . . . . . . . . . . 572
19.3.1 Definition of Behavioral Rules . . . . . . . . . . . 57419.3.2 Precise Mobility Modeling of Individuals . . . . . 57419.3.3 Dynamic Interactions . . . . . . . . . . . . . . . . 574
19.4 Modeling of Movement from Behavioral Rules . . . . . . . . 57519.5 BM Model for Individual Pedestrian Mobility . . . . . . . . 57619.6 BM Model Group Mobility . . . . . . . . . . . . . . . . . . 578
19.6.1 Evaluation . . . . . . . . . . . . . . . . . . . . . . 58019.7 Practical Issues and Tradeoffs . . . . . . . . . . . . . . . . . 581
19.7.1 Computational Complexity . . . . . . . . . . . . . 58119.7.2 Practical Issues . . . . . . . . . . . . . . . . . . . 58119.7.3 Benefits of BM Modeling Approach to Mobility . . 582
xx Contents
19.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58319.9 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 583References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584
20 Steady-State Generic Mobility . . . . . . . . . . . . . . . . . . . . 58520.1 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58520.2 Random Trip Mobility Model . . . . . . . . . . . . . . . . . 586
20.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . 58620.3 Random Trip Mobility . . . . . . . . . . . . . . . . . . . . . 587
20.3.1 Model Description . . . . . . . . . . . . . . . . . 58720.3.2 Strong Stochastic Stability for Random Trip . . . . 59020.3.3 Summary . . . . . . . . . . . . . . . . . . . . . . 593
20.4 Clustered Mobility Model . . . . . . . . . . . . . . . . . . . 59320.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . 59320.4.2 Characteristics of Scale-Free Mobile Ad
Hoc Network . . . . . . . . . . . . . . . . . . . . 59420.5 Clustered Mobility Model Description . . . . . . . . . . . . 595
20.5.1 Analysis of CMM . . . . . . . . . . . . . . . . . . 59720.5.2 Performance Evaluation . . . . . . . . . . . . . . . 60020.5.3 Summary . . . . . . . . . . . . . . . . . . . . . . 605
20.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 606References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 606
21 Graph-Based Mobility . . . . . . . . . . . . . . . . . . . . . . . . 60721.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 60721.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60821.3 Graph Walk Mobility Model . . . . . . . . . . . . . . . . . . 608
21.3.1 Routing Protocols . . . . . . . . . . . . . . . . . . 61121.3.2 Simulation Environment . . . . . . . . . . . . . . 61321.3.3 Simulation Results . . . . . . . . . . . . . . . . . 61521.3.4 Summary . . . . . . . . . . . . . . . . . . . . . . 620
21.4 Area Graph-Based Mobility Model . . . . . . . . . . . . . . 62121.4.1 Broadcast Protocols . . . . . . . . . . . . . . . . . 62321.4.2 Experimental Studies . . . . . . . . . . . . . . . . 62521.4.3 Summary . . . . . . . . . . . . . . . . . . . . . . 632
21.5 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 633References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 633
Part III Group Mobility Models
22 Reference Point Group Mobility . . . . . . . . . . . . . . . . . . . 63722.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 63722.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63722.3 RPGM Model Description . . . . . . . . . . . . . . . . . . . 63822.4 Applications of RPGM Model . . . . . . . . . . . . . . . . . 64222.5 Modified Version of RPGM Model . . . . . . . . . . . . . . 643
Contents xxi
22.6 Performance with RPGM Model . . . . . . . . . . . . . . . 64422.6.1 Performance Metrics . . . . . . . . . . . . . . . . 64422.6.2 Simulation Environment . . . . . . . . . . . . . . 64522.6.3 Simulation Results . . . . . . . . . . . . . . . . . 646
22.7 Exponential Correlated Random Group Mobility . . . . . . . 65122.8 Column Mobility . . . . . . . . . . . . . . . . . . . . . . . . 65222.9 Nomadic Community Mobility . . . . . . . . . . . . . . . . 654
22.9.1 Topology Control Algorithm UsingNomadic Community Mobility . . . . . . . . . . . 655
22.10 Pursue Mobility . . . . . . . . . . . . . . . . . . . . . . . . 66422.10.1 Performance Evaluation . . . . . . . . . . . . . . . 666
22.11 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66822.12 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 669References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 670
23 Reference Velocity Group Mobility . . . . . . . . . . . . . . . . . 67123.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 67123.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67223.3 RVGM Model Description . . . . . . . . . . . . . . . . . . . 67323.4 RVGM Model Applications . . . . . . . . . . . . . . . . . . 676
23.4.1 Mobile Ad Hoc Network Partitioning Problems . . 67623.4.2 Partition Prediction . . . . . . . . . . . . . . . . . 67723.4.3 Partition Prediction Algorithm . . . . . . . . . . . 67723.4.4 Application of Partition Prediction . . . . . . . . . 68023.4.5 Mobile Node Velocity Clustering . . . . . . . . . . 68123.4.6 Sequential Clustering Algorithm . . . . . . . . . . 68123.4.7 Illustration of Sequential Clustering Algorithm . . 682
23.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68323.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 684References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 684
24 Reference Velocity and Acceleration Group Mobility . . . . . . . 68524.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 68524.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68624.3 RVAG Mobility Model Description . . . . . . . . . . . . . . 68624.4 Clustering Algorithm . . . . . . . . . . . . . . . . . . . . . 68824.5 Partition Prediction Scheme . . . . . . . . . . . . . . . . . . 68924.6 Performance Evaluation . . . . . . . . . . . . . . . . . . . . 69024.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69324.8 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694
25 Structured Group Mobility . . . . . . . . . . . . . . . . . . . . . . 69525.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 69525.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69625.3 Structured Group Mobility Model Description . . . . . . . . 696
xxii Contents
25.4 Applications of Structured Group Mobility Model . . . . . . 69925.4.1 Firefighters Operating in a Building . . . . . . . . 69925.4.2 Military Units on the Battlefield . . . . . . . . . . 699
25.5 MANET Behavior in Face of Induced Link Breakages . . . . 70025.6 Simulations with Structured Groups . . . . . . . . . . . . . . 701
25.6.1 Simulator . . . . . . . . . . . . . . . . . . . . . . 70125.6.2 Movement Patterns . . . . . . . . . . . . . . . . . 70225.6.3 Evaluation . . . . . . . . . . . . . . . . . . . . . . 703
25.7 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70525.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70925.9 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 710References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 710
26 Virtual Track-Based Group Mobility . . . . . . . . . . . . . . . . 71126.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 71126.2 Virtual Track-Based Group Mobility Description . . . . . . . 71226.3 Defining Switch Stations and Virtual Tracks . . . . . . . . . 71326.4 Initial Node Distribution and Group Affiliation . . . . . . . . 71426.5 Group Mobility Under Constraint of Tracks . . . . . . . . . . 71426.6 Group Split/Merge at the Switch Station . . . . . . . . . . . 71426.7 Random and Individual Nodes Mobility . . . . . . . . . . . . 71526.8 Simulation Evaluation . . . . . . . . . . . . . . . . . . . . . 715
26.8.1 Simulation Platform . . . . . . . . . . . . . . . . . 71526.8.2 Performance with Mobile Groups . . . . . . . . . . 71526.8.3 Impact of Individual Random Moving and
Static Nodes . . . . . . . . . . . . . . . . . . . . . 71826.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72026.10 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 720References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 720
27 Drift Group Mobility . . . . . . . . . . . . . . . . . . . . . . . . . 72127.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 72127.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72227.3 Drift Group Mobility Description . . . . . . . . . . . . . . . 72327.4 Applications of Drift Group Mobility on Routing Group . . . 72527.5 Performance Results . . . . . . . . . . . . . . . . . . . . . . 72827.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73127.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 732References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 732
28 Gathering Group Mobility . . . . . . . . . . . . . . . . . . . . . . 73328.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 73328.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73428.3 Scale-Free Characteristics in Mobility . . . . . . . . . . . . . 73428.4 Gathering Group Mobility Model Description . . . . . . . . 73728.5 Experiment Results . . . . . . . . . . . . . . . . . . . . . . 738
Contents xxiii
28.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74028.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 740References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 740
29 Group Force Mobility . . . . . . . . . . . . . . . . . . . . . . . . . 74329.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 74329.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74529.3 Group Force Mobility Model Description . . . . . . . . . . . 746
29.3.1 Basic Force Model . . . . . . . . . . . . . . . . . 74629.3.2 Group Force Mobility Model . . . . . . . . . . . . 747
29.4 Simulation and Results . . . . . . . . . . . . . . . . . . . . 75029.4.1 Simulation Methodology . . . . . . . . . . . . . . 75029.4.2 Simulation Results . . . . . . . . . . . . . . . . . 752
29.5 Performance Assessment . . . . . . . . . . . . . . . . . . . 75429.5.1 Performance Metrics and Methodology . . . . . . . 75429.5.2 Performance Analysis . . . . . . . . . . . . . . . . 755
29.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75829.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 758References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 759
30 Group Mobility Extending Individual Mobility Models . . . . . . 76130.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 76130.2 Group Mobility Models . . . . . . . . . . . . . . . . . . . . 76230.3 Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . 76430.4 Basic Scenarios . . . . . . . . . . . . . . . . . . . . . . . . 76530.5 Impact of Node Speeds . . . . . . . . . . . . . . . . . . . . 76730.6 Impact of Group Numbers . . . . . . . . . . . . . . . . . . . 76930.7 Impact of Area Size . . . . . . . . . . . . . . . . . . . . . . 77030.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77130.9 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 771References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 772
Part IV Autoregressive Mobility Models
31 Autoregressive Individual Mobility . . . . . . . . . . . . . . . . . 77531.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 77531.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77631.3 Individual Linear System Mobility Model . . . . . . . . . . 77731.4 Autoregressive Individual Mobility Model Description . . . . 77831.5 Observation Data . . . . . . . . . . . . . . . . . . . . . . . 78031.6 Mobility Tracking Algorithm . . . . . . . . . . . . . . . . . 782
31.6.1 Pre-filtering . . . . . . . . . . . . . . . . . . . . . 78331.6.2 Initialization Module . . . . . . . . . . . . . . . . 78431.6.3 Extended Kalman Filter for Mobility State
Estimation . . . . . . . . . . . . . . . . . . . . . . 78431.6.4 AMM with First-Order Autocorrelation
of Mobility . . . . . . . . . . . . . . . . . . . . . 785
xxiv Contents
31.7 Numerical Results . . . . . . . . . . . . . . . . . . . . . . . 78531.7.1 Simulation Setup . . . . . . . . . . . . . . . . . . 78631.7.2 Mobility Estimation and Prediction . . . . . . . . . 786
31.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78831.9 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789
32 Autoregressive Group Mobility Model . . . . . . . . . . . . . . . . 79132.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 79132.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79232.3 Autoregressive Group Mobility Model Description . . . . . . 79332.4 Detection and Estimation of Group Mobility . . . . . . . . . 795
32.4.1 Group Mobility Detection – CorrelationIndex Test . . . . . . . . . . . . . . . . . . . . . . 795
32.4.2 Group Mobility Estimation . . . . . . . . . . . . . 79632.5 Numerical Results . . . . . . . . . . . . . . . . . . . . . . . 797
32.5.1 GPS Data . . . . . . . . . . . . . . . . . . . . . . 79732.5.2 Simulation Data . . . . . . . . . . . . . . . . . . . 802
32.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80532.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 805References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 806
Part V Flocking/Swarming Mobility Models
33 Flocking Mobility Models . . . . . . . . . . . . . . . . . . . . . . 80933.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 80933.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81033.3 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . 810
33.3.1 Graphs and Nets . . . . . . . . . . . . . . . . . . . 81233.3.2 α-Lattices and Quasi-α-Lattices . . . . . . . . . . 81333.3.3 Deviation Energy of Conformation . . . . . . . . . 81333.3.4 α-Norms and Bump Functions . . . . . . . . . . . 81433.3.5 Collective Potential Functions . . . . . . . . . . . 815
33.4 Flocking Algorithm for Free Space . . . . . . . . . . . . . . 81733.4.1 Collective Dynamics . . . . . . . . . . . . . . . . 81833.4.2 Stability Analysis of Flocking in Free Space . . . . 819
33.5 Flocking with Obstacle Avoidance . . . . . . . . . . . . . . 82033.5.1 β-Neighbors of α-Agents and (α, β)-Nets . . . . . 82133.5.2 Constrained α-Lattices . . . . . . . . . . . . . . . 82233.5.3 Multi-species Collective Potentials . . . . . . . . . 82233.5.4 Flocking Algorithm in the Presence of Obstacles . . 82333.5.5 Calculation of Position and Velocity of β-Agents . 82433.5.6 Analysis of Flocking with Obstacle Avoidance . . . 825
33.6 Flocking Mobility Model for Mobile Ad Hoc Networks . . . 82633.6.1 Flocking-Based Mobility Model for Ad
Hoc Network in Arbitrary m-Dimensional Space . . 828
Contents xxv
33.7 Peer-to-Peer Information Flow with Constrained Flocking . . 82933.8 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . 830
33.8.1 Two-Dimensional Flocking in Free Space . . . . . 83133.8.2 Two-Dimensional Fragmentation in Free Space . . 83233.8.3 Three-Dimensional Flocking in Free Space . . . . 83333.8.4 Split and Rejoin Maneuver . . . . . . . . . . . . . 83433.8.5 Squeezing Maneuver: Moving Through
Narrow Spaces . . . . . . . . . . . . . . . . . . . 83633.9 What Constitutes Flocking? . . . . . . . . . . . . . . . . . . 838
33.9.1 Verification of α-Flocking . . . . . . . . . . . . . . 83833.10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83933.11 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 840References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 841
34 Swarm Group Mobility Model . . . . . . . . . . . . . . . . . . . . 84334.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 84334.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84434.3 Swarm Group Mobility Model Description . . . . . . . . . . 844
34.3.1 Physical Model . . . . . . . . . . . . . . . . . . . 84534.3.2 Perception Model . . . . . . . . . . . . . . . . . . 84534.3.3 Behavioral Model . . . . . . . . . . . . . . . . . . 84734.3.4 Complexity . . . . . . . . . . . . . . . . . . . . . 849
34.4 Experimentation . . . . . . . . . . . . . . . . . . . . . . . . 84934.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85234.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 852References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 852
Part VI Virtual Game-Driven Mobility Models
35 Virtual Game-Driven Mobility Models Description . . . . . . . . 85735.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 85735.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85735.3 Virtual Game-Driven Mobility Framework
and Abstraction . . . . . . . . . . . . . . . . . . . . . . . . 85835.4 User Movements . . . . . . . . . . . . . . . . . . . . . . . . 86035.5 Analysis of the Ad Hoc Network Performance
Data Using Multi-player Game . . . . . . . . . . . . . . . . 86235.6 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . 86635.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87435.8 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 875References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 875
Part VII Non-recurrent Mobility Models
36 Non-recurrent Mobility Models . . . . . . . . . . . . . . . . . . . 87936.1 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 879
xxvi Contents
36.2 KDS-Based Mobility Model Preliminaries . . . . . . . . . . 88036.2.1 Definitions . . . . . . . . . . . . . . . . . . . . . . 88036.2.2 KDS Framework . . . . . . . . . . . . . . . . . . 88436.2.3 Motion in Computational Geometry . . . . . . . . 88536.2.4 Kinetic Data Structures . . . . . . . . . . . . . . . 88736.2.5 Kinetic Sorted List . . . . . . . . . . . . . . . . . 88836.2.6 General Polygonal Models . . . . . . . . . . . . . 88936.2.7 Spline and Algebric Objects . . . . . . . . . . . . 89236.2.8 Dynamic Queries . . . . . . . . . . . . . . . . . . 89336.2.9 Proximity Maintenance . . . . . . . . . . . . . . . 893
36.3 Mobility Model Formulation in KDS . . . . . . . . . . . . . 89436.3.1 Swept Volume-Based Collision Detection . . . . . 89436.3.2 Methodology for Solving Problems . . . . . . . . . 89636.3.3 Motion Formulation . . . . . . . . . . . . . . . . . 89736.3.4 Boundary Volume Hierarchies Generation
and Culling . . . . . . . . . . . . . . . . . . . . . 89936.3.5 Swept Volume Generation . . . . . . . . . . . . . 90036.3.6 Collision Detection in KDS . . . . . . . . . . . . . 90136.3.7 Implementation . . . . . . . . . . . . . . . . . . . 90336.3.8 Main Results . . . . . . . . . . . . . . . . . . . . 905
36.4 KDS-Based Mobility Application to Mobile Ad HocNetworks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 906
36.5 Some Non-KDS-Based Mobility Models . . . . . . . . . . . 90836.5.1 Contraction Mobility Model . . . . . . . . . . . . 90836.5.2 Modified Contraction Mobility Model . . . . . . . 90936.5.3 Expansion Mobility Model . . . . . . . . . . . . . 909
36.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90936.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 910References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 910
Part VIII Social-Based Community Mobility Model
37 Time-Variant, Community-Based, and Home-CellCommunity-Based Mobility Model . . . . . . . . . . . . . . . . . 91337.1 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91337.2 Time-Variant Community Mobility Model . . . . . . . . . . 915
37.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . 91537.2.2 Theoretical Analysis of the TVC Mobility Model . 91737.2.3 TVC Mobility Model Validation . . . . . . . . . . 92737.2.4 Performance Prediction Using TVC Model . . . . . 93237.2.5 Summary . . . . . . . . . . . . . . . . . . . . . . 934
37.3 Community-Based Mobility Model . . . . . . . . . . . . . . 93437.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . 93437.3.2 CBM Description . . . . . . . . . . . . . . . . . . 93437.3.3 Gregorian Behavior in CBM . . . . . . . . . . . . 936
Contents xxvii
37.4 Home-Cell CBM Model . . . . . . . . . . . . . . . . . . . . 94337.4.1 HCBM Versus CBM in Controlling Node
Positions . . . . . . . . . . . . . . . . . . . . . . . 94437.4.2 Modified HCBM . . . . . . . . . . . . . . . . . . 94737.4.3 Modeling Movements in HCBM . . . . . . . . . . 95037.4.4 Mobility Pattern Evaluation . . . . . . . . . . . . . 95637.4.5 Summary . . . . . . . . . . . . . . . . . . . . . . 962
37.5 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 962References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 963
38 Orbit-Based Mobility . . . . . . . . . . . . . . . . . . . . . . . . . 96538.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 96538.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96638.3 Orbit-Based Mobility Parameters . . . . . . . . . . . . . . . 96738.4 General Orbit Mobility Model . . . . . . . . . . . . . . . . . 96838.5 Random Orbit Model . . . . . . . . . . . . . . . . . . . . . 96838.6 Uniform Orbit Model . . . . . . . . . . . . . . . . . . . . . 96938.7 Restricted Orbit Model . . . . . . . . . . . . . . . . . . . . 96938.8 Overlay Orbit Model . . . . . . . . . . . . . . . . . . . . . . 96938.9 Orbit-Aware Routing in Mobile Ad Hoc Network . . . . . . 970
38.9.1 Analytical Model . . . . . . . . . . . . . . . . . . 97138.9.2 Routing Protocol Description . . . . . . . . . . . . 97838.9.3 Performance Analysis . . . . . . . . . . . . . . . . 983
38.10 Comparison with Other Mobility Models . . . . . . . . . . . 98938.11 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99038.12 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 991References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 992
39 Entropy-Based Individual/Community Mobility Model . . . . . . 99339.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 99339.2 Notations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99339.3 Entropy-Based Mobility Model in Virtual Space . . . . . . . 995
39.3.1 Virtual Space Concept . . . . . . . . . . . . . . . . 99539.3.2 Mobility Pattern Characterization . . . . . . . . . . 99539.3.3 Entropy-Based Mobility Model Description . . . . 99939.3.4 Simulation Results . . . . . . . . . . . . . . . . . 99939.3.5 Summary . . . . . . . . . . . . . . . . . . . . . . 1005
39.4 Entropy-Based Mobility Model in CommunityStructure Environment . . . . . . . . . . . . . . . . . . . . . 100639.4.1 Mobility Model Description . . . . . . . . . . . . 100639.4.2 Applications in Mobile Ad Hoc Network . . . . . . 100839.4.3 Summary . . . . . . . . . . . . . . . . . . . . . . 1018
39.5 Entropy-Based Mobility in QOS and Clustering of MANET . 101939.5.1 Mobility Model Description . . . . . . . . . . . . 101939.5.2 Optimized Entropy-Based WCA with Tabu Search . 102339.5.3 Entropy-Based WCA Simulation Results . . . . . . 1025
xxviii Contents
39.5.4 Entropy-Based QOS Routing Simulation Results . 102939.5.5 Summary . . . . . . . . . . . . . . . . . . . . . . 1031
39.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1032References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1033
40 Knowledge-Driven Mobility Model . . . . . . . . . . . . . . . . . 103540.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 103540.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103640.3 KDM Model Description . . . . . . . . . . . . . . . . . . . 1037
40.3.1 Parameterizing Host Motion . . . . . . . . . . . . 103740.3.2 Exploiting Motion Profiles for Finding Services . . 103840.3.3 Service Requests by Mobile Hosts . . . . . . . . . 1039
40.4 Mobile Service Satisfying Set . . . . . . . . . . . . . . . . . 104040.4.1 Service Satisfiability . . . . . . . . . . . . . . . . 104040.4.2 Service Reachability . . . . . . . . . . . . . . . . 104140.4.3 Logical Service Mobility . . . . . . . . . . . . . . 1041
40.5 Exploiting KDM Model in Ad Hoc Mobile Environments . . 104340.5.1 Proactive Service Relocation . . . . . . . . . . . . 104340.5.2 Ensuring Continuous Connectivity . . . . . . . . . 1044
40.6 KDM Functional Architecture . . . . . . . . . . . . . . . . . 104540.6.1 Middleware Software Architecture . . . . . . . . . 104640.6.2 Knowledge Management System . . . . . . . . . . 104740.6.3 Knowledge Representation . . . . . . . . . . . . . 104840.6.4 Knowledge Base . . . . . . . . . . . . . . . . . . 105040.6.5 Knowledge Aggregation . . . . . . . . . . . . . . 105140.6.6 Knowledge Dissemination . . . . . . . . . . . . . 105340.6.7 Anatomy of Service Request . . . . . . . . . . . . 1053
40.7 KDM Simulation and Results . . . . . . . . . . . . . . . . . 105640.7.1 Simulation Setup . . . . . . . . . . . . . . . . . . 105640.7.2 Results Varying Networking Parameters . . . . . . 1057
40.8 Some Observations . . . . . . . . . . . . . . . . . . . . . . 106240.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106340.10 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1064References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1065
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1067
List of Figures
2.1 Movement of a node in a MANET: (a) epoch mobilityvectors and (b) resultant mobility vector [4].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 26
2.2 Joint mobility transformation: (a) joint node case and(b) joint mobility transformation [4]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.1 Traveling pattern of an MN using the 2D randomwalk mobility model (time) [3]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 36
3.2 Traveling pattern of an MN using the 2D randomwalk mobility model (distance) [3]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 37
3.3 Generalized mobility model simulation algorithm [12] . . . . . . . 523.4 Marginal X-coordinate steady-state probability for random
walk (W = 135 m) [12] . . . . . . . . . . . . . . . . . . . . . . . 533.5 Chi-squared distance for steady-state location
approximation [12] . . . . . . . . . . . . . . . . . . . . . . . . . 543.6 Markov chain and probability matrix of Markovian random
walk mobility model [3]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
3.7 Mobile node’s movement pattern using Markovian randomwalk model [3]. © IEEE – Reproduced with permission . . . . . . 59
3.8 Markov chain state transition for simple individual mobilityMarkovian model [22] . . . . . . . . . . . . . . . . . . . . . . . . 59
4.1 Traveling pattern of a mobile node using the randomwaypoint mobility model [2]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.2 Average neighbor in percentage versus time [2].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 66
4.3 PDF of transition length of RWP nodes in a rectangle [1].© Kluwer Academic Publishers – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
xxix
xxx List of Figures
4.4 Expected length and PDF: (a) expected transition lengthof RWP nodes within an a × b rectangle and (b) PDF oftransition length of RWP nodes on a disk of radius a [1].© Kluwer Academic Publishers – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.5 Spatial node distribution resulting from the RWPmodel (simulation results): (a) square system areaand (b) circular system area [1]. © Kluwer AcademicPublishers – Reproduced with permission . . . . . . . . . . . . . 80
4.6 Definition of direction angles [1]. © Kluwer AcademicPublishers – Reproduced with permission . . . . . . . . . . . . . 83
4.7 Distribution of movement direction: f� (θ |r), r/a, and θ [1].© Kluwer Academic Publishers – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
4.8 Distribution of movement direction: f� (θ) and θ [1].© Kluwer Academic Publishers – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
4.9 Cell changes per transition [1]. © Kluwer AcademicPublishers – Reproduced with permission . . . . . . . . . . . . . 87
4.10 Expected cell boundary change rate E {Ct} on a square areaof size ‖A‖ = 62, 500α2 m2, Number of cells = α2, andlength of one square cell = 250 m [1]. © Kluwer AcademicPublishers – Reproduced with permission . . . . . . . . . . . . . 89
4.11 Random waypoint on a non-convex Swiss flag domain [14].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 95
4.12 Random waypoint on a non-convex city sectiondomain [14]. © IEEE – Reproduced with permission . . . . . . . 95
4.13 Restricted random waypoint on a plane with foursquares [14]. © IEEE – Reproduced with permission . . . . . . . 96
4.14 Restricted random waypoint fish-in-a-bowl surface [14].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 97
4.15 Random waypoint on a sphere [14]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
4.16 Random waypoint (or random walk) with wrapping [14].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 99
4.17 Random waypoint (or random walk) with billiard-likereflection at the edges of the domain [14]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 100
4.18 Markov model of location transition of mobile nodes . . . . . . . 1024.19 Zigzag movement of the RWP process [19]. © IEEE –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1044.20 Illustration of the variables: P1, φ, r, �, dA, and a1 [19].
© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 1064.21 Illustration of the integral over [0, 2π ] [19].
© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 106
List of Figures xxxi
4.22 Derivation of a1 and a2 in a unit disk [19].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 107
4.23 The pdf of the node location, f(r), (left) and the pdf ofthe distance of a node from the origin, fR(r), (right) fora unit disk. The solid curves correspond to our exactresults and the dashed curves to approximation P1(r)(see Table 4.4) [19]. © IEEE – Reproduced with permission . . . . 108
4.24 Notation for analysis of RWPB [19].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 111
4.25 The pdf is resulting from the RWPB model in unit square.The left figure corresponds to pdf of the interior mode f0(r),and the right figure corresponds to pdf of the border modefi(r) [19]. © IEEE – Reproduced with permission . . . . . . . . . 114
4.26 Notation for analysis of RWPB in a unit circle [19].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 115
4.27 The cdf FR(r) of the distance of the node from the origin(left) and the pdf f (r) = f (|r|) of the node location(middle and right) for the RWPB model in a unit disk [19].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 116
4.28 Comparison of Cn(d) with RWP node distribution(dashed lines), RWPB node distribution (solid lines),and uniform node distribution (dotted lines) for n= 20,n= 100, and n= 500 nodes (from left to right) [19].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 117
4.29 On the left figure the lower curve corresponds to the pdf ofthe node location according to RWP model and the uppercurve the pdf of the packet location in a dense ad hocnetwork. The figures on the right illustrate the respectivepdfs in three-dimensions [19]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.1 Speed behavior v(t) of car in downtown [6].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 129
5.2 Direction behavior of a node using smooth randombehavior [6]. © ACM – Reproduced with permission . . . . . . . 131
5.3 Three mobility traces [6]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.4 Speed slowdown and node distribution: (a) modeling ofslowdown of mobile nodes before turning and (b) spatialnode distribution histogram [6]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
5.5 The first node enters the transmission range of the secondnode at an angle φ to the tangent at A and moves along thechord AB [7]. © Inderscience – Reproduced with permission . . . 139
5.6 Comparison of theoretical and simulations results for theexpected hitting and meeting times under the random
xxxii List of Figures
direction (or smooth random) mobility model [7].© Inderscience – Reproduced with permission . . . . . . . . . . . 150
5.7 Random direction (or smooth random) mobility model:(a) comparison of the theoretical and simulation results forthe expected contact time for parameters N = 100 × 100,T = 300, v = 1. (b) meeting time distribution withparameters N = 100 × 100, K = 30, T = 160, v = 1,Tstop = 150. (c) Inter-meeting time distribution withparameters N = 100 × 100, K = 30, T = 160, v = 1,Tstop = 150 [7]. © Inderscience – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
5.8 Node A will pass through the transmission range of node Bif and only if its destination lies in the shaded region [7].© Inderscience – Reproduced with permission . . . . . . . . . . . 155
5.9 Comparison of theoretical and simulation results for theexpected hitting and meeting times under the randomwaypoint model [7]. © Inderscience – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
5.10 Random waypoint mobility model: (a) comparison ofthe theoretical and simulation results for the expectedcontact time for parameters N = 100 × 100, v = 1,Tstop = 150; (b) meeting time distribution with parametersN = 100 × 100, K = 30, v = 1, Tstop = 150; and(c) inter-meeting time distribution with parametersN = 100× 100, K = 30, T = 160, v = 1, Tstop = 150 [7].© Inderscience – Reproduced with permission . . . . . . . . . . . 158
5.11 Upper and lower bounds on the delay ofmobility-assisted routing scheme under randomdirection [7]. © Inderscience – Reproduced with permission . . . . 160
5.12 Simulation and analytical results for the expected delay of(a) direct transmission and (b) epidemic routing. Networkparameters: N = 150 × 150, v = 1, L = 55, M = 50,T = 0 [7]. © Inderscience – Reproduced with permission . . . . . 163
6.1 Freeway mobility model: (a) freeway mobility model,(b) mobile node traveling pattern in freeway/city section,and (c) pathway graph used in freeway . . . . . . . . . . . . . . . 170
6.2 Mobile nodes (e.g., vehicles) moving in two-lane freeway[5]. © IEEE – Reproduced with permission . . . . . . . . . . . . 171
6.3 Mobility rules in multi-lane freeways for mobile nodes . . . . . . 1726.4 Manhattan mobility model . . . . . . . . . . . . . . . . . . . . . 1736.5 Car-following mobility model algorithm . . . . . . . . . . . . . . 1756.6 Evolution of speed and headway time for the first 20
vehicles belonging to a queue of cars meeting a slowvehicle ahead (at time t = 60 s, the slow vehicle startsaccelerating) [5]: (a) freeway model, (b) FTM model,
List of Figures xxxiii
(c) IDM model, and (d) Kraub model. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 179
6.7 Speed waves generated by the IDM model in presence ofsevere traffic congestions on the highway scenario [5].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 179
6.8 Average speed profile of vehicular outflow (top) andinflow (bottom) in presence of an intersection [5].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 180
6.9 Vehicular density in an urban scenario obtained withthe IDM-IM model [5]: (a) random mobility and(b) activity-based mobility. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
6.10 PDR distribution over hop count with random mobility(top) and activity-based mobility (bottom) [5]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 183
6.11 PDR versus time for different configurations [5] (a) singles/d pair, random mobility, (b) single s/d pair, activity-basedmobility, and (c) multiple s/d pairs, activity-based mobility,IDM-IM model. © IEEE – Reproduced with permission . . . . . . 184
6.12 Voronoi diagram with obstacles and terrain with labeledsites [3]. © IEEE – Reproduced with permission . . . . . . . . . . 187
6.13 Voronoi diagram for terrain with labeled sites [3]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 187
6.14 Example of semi-definite nodemovement [3]. © IEEE – Reproduced with permission . . . . . . . 189
6.15 Example network scenario for simulation by obstaclemobility model [3]. © IEEE – Reproduced with permission . . . . 191
6.16 Node density [3]. © IEEE – Reproduced with permission . . . . . 1956.17 Path length versus time [3]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1966.18 Average link duration [3]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1976.19 Data packet reception [3]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1976.20 Control packet overhead [3]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1986.21 End-to-end latency [3]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1996.22 Data packet reception versus signal attenuation
percentage [3]. © IEEE – Reproduced with permission . . . . . . 2006.23 Data packet reception [3]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2016.24 Control packet overhead [3]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
xxxiv List of Figures
6.25 End-to-end latency [3]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
6.26 Movement of a node in community structure [10].© Academic Publisher – Reproduced with permission . . . . . . . 207
6.27 Simulated terrain . . . . . . . . . . . . . . . . . . . . . . . . . . 2106.28 Average node density versus simulation time for the
campus terrain [10]. © Academic Publisher – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
6.29 Percentage node connectivity versus simulation timefor different scenarios [10]. © Academic Publisher –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 211
6.30 Voronoi path of the obstacle mobility model [11]. © VDEVerlag – Reproduced with permission . . . . . . . . . . . . . . . 214
6.31 First-order Voronoi graph with building interpolation depthx = 3 [11]. © VDE Verlag – Reproduced with permission . . . . . 215
6.32 First-order Voronoi graph after refinement [11]. © VDEVerlag – Reproduced with permission . . . . . . . . . . . . . . . 216
6.33 Second-order Voronoi graph with building interpolationdepth x = 3 [11]. © VDE Verlag – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
6.34 Second-order Voronoi graph after refinement [11]. © VDEVerlag – Reproduced with permission . . . . . . . . . . . . . . . 217
6.35 Positioning of a mobile node [11]. © VDE Verlag –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 218
6.36 Illustration of the Voronoi mobility model for three givennodes [11]. © VDE Verlag – Reproduced with permission . . . . . 220
7.1 A mobile node’s movement pattern using random directionmobility [5]. © IEEE – Reproduced with permission . . . . . . . . 224
7.2 Walk segment definition of a random direction walkpolygon [7]. © COST – Reproduced with permission . . . . . . . 225
7.3 Sojourn density (top) and movement vector field (middle)of a RDM walker with gravity and five mass points(bottom) and 80 kg walker mass for RDM-LDP model [7].© COST – Reproduced with permission . . . . . . . . . . . . . . 228
7.4 Grid of a RDM-LDP model with location-dependentparameter [7]. © COST – Reproduced with permission . . . . . . 229
7.5 Sojourn density distribution of a RDM-LDP walkerwith prolonged walk segment length distributions for theRDM-LDP grid fields (4, 0)−(4, 3) of a 10× 8 grid [7].© COST – Reproduced with permission . . . . . . . . . . . . . . 230
7.6 Sojourn density plot (top) and movement vector field(bottom) of a RDM-LDP walker with two directedlanes at the fields (4, 1)–(4, 6) and (5, 1)–(5, 6) of a10×8 RDM-LDP grid [7]. © COST – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
List of Figures xxxv
7.7 Process for RDM-LDM parameterization . . . . . . . . . . . . . . 2327.8 Graphical TOOL to edit 2D arrays of empirical distributions
[7]. © COST – Reproduced with permission . . . . . . . . . . . . 2327.9 Example procedures for conversion from RWP to
RDM-LDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2337.10 Normalized error between RWP and RDM-LDP [7].
© COST – Reproduced with permission . . . . . . . . . . . . . . 2347.11 Average neighbors per node at 1 m/s mobility [6]. © IEEE –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . 2377.12 Average neighbor per node at 5 m/s mobility [6]. © IEEE –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . 2387.13 Simulated movement in the random direction mobility
model [6]. © IEEE – Reproduced with permission . . . . . . . . . 2397.14 Number of node pairs at a given distance . . . . . . . . . . . . . . 2397.15 Number of packets delivered [6]. © IEEE – Reproduced
with permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 2407.16 Normalized throughput [6]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2417.17 Average path length [6]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2427.18 Probability of being able to establish initial route [6].
© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 2438.1 Deterministic motion of nodes . . . . . . . . . . . . . . . . . . . 2478.2 Global mobility model – DMM with known mobility
patterns [3]. © IEEE – Reproduced with permission . . . . . . . . 2498.3 Mobility prediction [3]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2518.4 Cellular network environment [12]. © Tech-Net
Publishing – Reproduced with permission . . . . . . . . . . . . . 2588.5 Cell type transition probability matrix . . . . . . . . . . . . . . . 2598.6 Mobility trajectory toward attraction point A [12].
© Tech-Net Publishing – Reproduced with permission . . . . . . . 2608.7 City attracted: (a) 30 and (b) 60 min [12]. © Tech-Net
Publishing – Reproduced with permission . . . . . . . . . . . . . 2608.8 City attracted: (a) 180 and (b) 210 min [12]. © Tech-Net
Publishing – Reproduced with permission . . . . . . . . . . . . . 2618.9 City attracted: (a) 530 and (b) 560 min [12]. © Tech-Net
Publishing – Reproduced with permission . . . . . . . . . . . . . 2618.10 City attracted: (a) 690 and (b) 710 min [12]. © Tech-Net
Publishing – Reproduced with permission . . . . . . . . . . . . . 2618.11 Handover traffic rate: (a) cell A and (b) cell B [12].
© Tech-Net Publishing – Reproduced with permission . . . . . . . 2629.1 Mobile node’s probabilistic direction of motion in partially
deterministic mobility model . . . . . . . . . . . . . . . . . . . . 268
xxxvi List of Figures
9.2 Conceptual system view for modeling of local mobilitymodel [2]. © IEEE – Reproduced with permission . . . . . . . . . 270
9.3 Cell geometry [2]. © IEEE – Reproduced with permission . . . . . 2809.4 A practical situation necessitates looking-ahead mode
for UMP identification [2]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
9.5 Actual and predicted user trajectory [2]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 282
9.6 Actual and predicted user trajectories with multipleUMPs [2]. © IEEE – Reproduced with permission . . . . . . . . . 283
9.7 Actual and predicted user speed [2]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
9.8 Matrices used for approximate pattern matching for cell c9with c10 as the local prediction of the next crossing cell [2].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 286
9.9 Parameter behavior of probability of correctness for localprediction [2]. © IEEE – Reproduced with permission . . . . . . . 288
9.10 Comparison of correctness of the next crossing cell usingHLP and MMP [2]. © IEEE – Reproduced with permission . . . . 289
9.11 Points A, B, and C are local minima on H(X), and point Cis the only global minimum [9]. © University of Waterloo –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 294
9.12 Given a neighborhood (gray areas), an element suchas site A is conditionally independent of the rest of thesystem. Changing the value of site B, for instance, doesnot directly affect site A. Regions such as C can alsobe conditionally independent, given an appropriate setof boundary elements [9]. © University of Waterloo –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 295
9.13 Simulation annealing algorithm [11]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
9.14 Total transmission power over time for different fixed β [8].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 303
9.15 Total transmission power when β is increaseddramatically [8]. © IEEE – Reproduced with permission . . . . . 304
9.16 Initial and final node positions [8]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
9.17 Mobility and transmission energy [8]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 304
9.18 Total transmission power over time [8].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 305
10.1 Change of mean angle near the edges (in degrees) [6].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 315
List of Figures xxxvii
10.2 Movement pattern of a mobile node using the randomGauss–Markov mobility model [6]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
10.3 Predictive location updating and selective paging: (a) 1Dsystem and (b) 2D system [4]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
10.4 Joint optimization of m and N versus β and σ , for λ = 0.01,Cu = 10, and Ci = 1: (a) minimum cost and (b) optimal mand N [4]. © IEEE – Reproduced with permission . . . . . . . . . 333
10.5 Joint optimization of m and N versus β and Ci, forβ = 10−0.5, σ = 0.5, and Cu = 10: (a) minimum costand (b) optimal m and N [4]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
10.6 Joint optimization of m and N versus β and Ci, forβ = 10−0.5, σ = 0.5, and Cu = 1: (a) minimum costand (b) optimal m and N [4]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
10.7 Performance gain with ideal random Gauss–Markovmobility model, for λ = 0.01, Cu = 10, and m = 10: (a)versus β and μ, with σ = 0.5, (b) versus μ and σ , withβ = 10−0.05(α = 0.73), and (c) versus σ and β, withμ = 0.5 [4]. © IEEE – Reproduced with permission . . . . . . . . 337
10.8 Performance gain with ideal random Gauss–Markovmobility model, for μ = 0.5, σ = 0.5, and β = 10−0.05: (a)versus λ and Cu, with m = 10, (b) versus Cu and m, withλ = 0.01, and (c) versus m and λ, with Cu = 10−1.5 [4].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 339
10.9 Dynamic parameter estimation with random waypointmobility model for Cu = 10: (a) comparison of minimum(over N) cost and (b) performance gain [4]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 341
11.1 An example of speed versus time in one SMSmovement [1]. © IEEE – Reproduced with permission . . . . . . . 348
11.2 Average speed versus simulation time [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 353
11.3 Two-dimensional spatial node distribution with1000 nodes: (a) RWP 2D at the 1000th s,(b) SMS 2D at the 500th s, and (c) SMS 2D at the1000th s [1]. © IEEE – Reproduced with permission . . . . . . . . 354
11.4 Link lifetime distribution and average node degreecomparison among RWP, RGM, and SMS models [1]:(a) PMF of link lifetime and (b) percentage of node degree.© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 355
11.5 Distance and trace length [11]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
xxxviii List of Figures
11.6 Routing performance comparison between RWP and SMSmodels: (a) average end-to-end packet delay, (b) end-to-endnetwork throughput, and (c) routing overhead ration [11].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 358
11.7 Network connectivity performance comparison betweenRWP and SMS models: (a) PDF of relative speed, (b) CDFof link lifetime, and (c) node degree [11]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 359
11.8 Group mobility based on SMS model [11]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 361
11.9 SMS model application in Manhattan-like city map [11].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 362
11.10 A sample of a relative movement trajectory between a nodepair (u, w) during their link connection [15]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 364
11.11 Distance transition probability matrix P [15]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 365
11.12 Comparison of the experimental relative speed distributionwith the Rayleigh distribution approximation [15]. ©IEEE – Reproduced with permission . . . . . . . . . . . . . . . . 368
11.13 Comparison of expected link lifetime between simulationresults and theoretical results according to vα = 2, 5, 10,15, and 20 m/s [15]. © IEEE – Reproduced with permission . . . . 370
11.14 Derivation of expected new link arrival rate λ [15]. ©IEEE – Reproduced with permission . . . . . . . . . . . . . . . . 371
11.15 Comparison of expected new link arrival rate λ andexpected link change rate ηl between simulation resultsand theoretical results, according to different node targetspeeds vα , where σ = 5/πR2, R = 250 m [15]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 372
11.16 Expected number of neighbors per node accordingto different node target speeds vα and node densityσ = number/πR2, where R = 250 m [15]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 374
11.17 Transmission range of node 1 at O with node 2 entering therange at A and exiting at B . . . . . . . . . . . . . . . . . . . . . 376
12.1 Coverage area mapped into a closed surface [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 381
12.2 Mobile node’s movement pattern in boundless simulationarea mobility model [3]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382
12.3 Process flow for mobility principle used in simulation . . . . . . . 38312.4 Example of zone routing [1]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
List of Figures xxxix
12.5 ZRP simulation results using BSA mobility model –comparison of the number of control messages [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 388
12.6 Node movements in mobile ad hoc network: (a) all nodesare moving and (b) only N0 is moving . . . . . . . . . . . . . . . 389
12.7 Motion of node Ni passing through the transmissionregion of another node N0 [4]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
12.8 Plot of pcomp versus normalized time τ [4]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 394
12.9 Examples of multihop routes – white nodes have multipleroutes to N0 while routes from black nodes are broken [4]:(a) all nodes have routes to N0 and (b) routes from N2 andN3 to N0 are broken. © IEEE – Reproduced with permission . . . 395
12.10 Simulation results with RWP mobility model [4]:(a) high-mobility case and (b) low-mobility case. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 396
12.11 Simulation results with RW mobility model [4]: (a) longmovetime case and (b) short movetime case. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 397
12.12 Simulation results with BSA mobility model [4]: (a) shorttransmission range case and (b) long transmission rangecase. © IEEE – Reproduced with permission . . . . . . . . . . . . 398
13.1 Greenshield’s relationship between the speed of traffic andits density [5]. © IPVS – Reproduced with permission . . . . . . . 409
13.2 Calculation of intra-cell outgoing rate in adaptivebeamforming system [7]. © University of Cincinnati –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 411
13.3 System model [6]. © Inderscience – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
13.4 Call admission procedures flowchart [6]. © Inderscience –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 414
13.5 Blocking probabilities versus traffic load [6].© Inderscience – Reproduced with permission . . . . . . . . . . . 422
13.6 Comparison of analytical model and simulation resultsfor data users [6]. © Inderscience – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
13.7 Comparison of analytical model and simulation resultsfor voice users [6]. © Inderscience – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424
13.8 Basic CSMA/CA protocol [10]. © Technology Interface –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 426
13.9 CSMA/CA with RTS/CTS protocol [10]. © TechnologyInterface – Reproduced with permission . . . . . . . . . . . . . . 426
xl List of Figures
13.10 Mobility model in MANET [10]. © Technology Interface –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 427
13.11 Structure of time Tv [10]. © Technology Interface –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 432
13.12 Multihop network model [10]. © Technology Interface –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 434
13.13 Illustration of hidden area [10]. © Technology Interface –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 434
13.14 Link breaking probability versus virtual packet transmissiontime in mobility model I [10]. © Technology Interface –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 438
13.15 Path breaking probability versus average moving speedin mobility model I [10]. © Technology Interface –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 439
13.16 Throughput of CSMA/CA protocols [10]. © TechnologyInterface – Reproduced with permission . . . . . . . . . . . . . . 440
14.1 Scalable mobility model system model [6]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 451
14.2 Overlay of PGs over city area of Bristol, UK, indicatingdifferent types of environment [6]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 452
14.3 Penetration matrix for city center of Bristol [6]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 456
14.4 Mobile user distribution over city center of Bristol, UK, at7:00 A.M. [6]. © ACM – Reproduced with permission . . . . . . . 457
14.5 General representation of stability conditions [6]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 458
14.6 Pole of gravity concept applied to characterize movementfor the city area model of Avon, Bristol, UK [6]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 461
14.7 Mobile user distribution over the city area of Avon, UK [6].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 462
14.8 Spatial distribution of subscribers at 7.00 A.M. [6].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 463
14.9 Spatial distribution of subscribers at 10.00 A.M. [6].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 463
14.10 System level simulator [6]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464
14.11 CDF of aggregate cell traffic distribution [6]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 465
14.12 CDF of handover [6]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466
14.13 Network snapshots at time 0, 200, 400, and 1400 s [7].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 476
List of Figures xli
14.14 Average node density [7]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477
14.15 Maximum node density [7]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478
14.16 Variance of node density [7]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478
14.17 Clustering coefficient [7]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
14.18 Network partitioning [7]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
15.1 Displacement measure [1]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 487
15.2 Link change versus mobility at different transmissionranges [1]. © ACM – Reproduced with permission . . . . . . . . . 488
15.3 Link change rate [1]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489
15.4 Link change rate using mobility vector model [1]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 489
15.5 Packet delivery ration for AODV [1]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 490
15.6 Packet delivery ration for DSR [1]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 491
15.7 Packet delivery ratio for FSR [1]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 491
16.1 Moving diagram of 2D-correlated random walk [1].© Middle East Technical University – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 498
16.2 Joint density function with Cx = Cy = 0.5, Dx = Dy = 1,and r = 0.7 [1]. © Middle East Technical University –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 504
16.3 Difference equation with boundary condition [1]. © MiddleEast Technical University – Reproduced with permission . . . . . 505
16.4 Resulting BMP [1]. © Middle East Technical University –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 506
16.5 Transformation procedure of BVP into its standardform [1]. © Middle East Technical University –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 507
16.6 BVP in circular-shaped domain [1]. © Middle EastTechnical University – Reproduced with permission . . . . . . . . 507
16.7 Hexagonal cell with motion starting at origin [1]. © MiddleEast Technical University – Reproduced with permission . . . . . 517
16.8 Rotation of axes of example given in Fig. 16.7 [1].© Middle East Technical University – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 518
xlii List of Figures
16.9 Resulting BVP from rotation of axes of example givenin Fig. 16.7 [1]. © Middle East Technical University –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 519
16.10 Resulting BVP from scaling step of example given inFig. 16.7 [1]. © Middle East Technical University –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 519
16.11 Solution plot for transformed BVP of example givenin Fig. 16.7 [1]. © Middle East Technical University –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 520
16.12 Top view of solution for transformed BVP of example givenin Fig. 16.7 [1]. © Middle East Technical University –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 520
16.13 Back transformation of solution into initial case [1].© Middle East Technical University – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521
16.14 Description of Example 2 [1]. © Middle East TechnicalUniversity – Reproduced with permission . . . . . . . . . . . . . 522
16.15 Solution plot for transformed BVP in Example 2 [1].© Middle East Technical University – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522
16.16 Top view of solution for transformed BVP in Example 2[1]. © Middle East Technical University – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 523
17.1 Panicking soccer fans [4]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534
17.2 Formation of lanes in initially disordered pedestrian crowdswith opposite walking directions [4]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 534
17.3 Oscillations of the passing direction at a bottleneck [4].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 536
17.4 Self-organized, short-lived roundabout traffic inintersecting pedestrian streams [4]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 537
17.5 Noise-induced formation of a crystallized, “frozen”state in a periodic corridor used by oppositely movingpedestrians [4]. © IEEE – Reproduced with permission . . . . . . 538
17.6 Simulation of pedestrians moving with identical desiredvelocity [4]. © IEEE – Reproduced with permission . . . . . . . . 539
17.7 Simulation of N=200 individuals fleeing from a linear firefront [4]. © IEEE – Reproduced with permission . . . . . . . . . . 540
17.8 Simulation of an escape route with a wider area [4].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 540
17.9 Simulation of N = 90 pedestrians trying toescape a smoky room of area A = 15 × 15 m(black) [4]. © IEEE – Reproduced with permission . . . . . . . . 542
List of Figures xliii
17.10 Coverage achieved when 100 nodes originally located incentral 10×10 m of a 150 × 150 m plane [6]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 544
17.11 Coverage achieved when 100 nodes originally located incentral 10×10 m of a 450×150 m plane – two guide nodesmove east and west from their initial positions in the centerof plane [6]. © IEEE – Reproduced with permission . . . . . . . . 548
17.12 Coverage achieved when 100 nodes originally located inbottom-left 10×10 m quadrant of a 110×110 m plane [6].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 550
17.13 Coverage achieved when 100 nodes originally located inbottom-left 10×10 m quadrant of a 110×110 m plane withone barrier [6]. © IEEE – Reproduced with permission . . . . . . 551
17.14 Coverage achieved when 100 nodes originally located inbottom-left 10×10 m quadrant of a 110×110 m plane withtwo barriers [6]. © IEEE – Reproduced with permission . . . . . . 552
17.15 Coverage achieved when 100 nodes originally located inbottom-left 10×10 m quadrant of a 110×110 m plane witha high, gently slopping hill located at (40,40) [6]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 553
18.1 Graphical representation of simulation area in HIMM [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 561
18.2 Graphical representation of the network influence matrixdivided into autonomous and dependent classes [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 563
18.3 Pedestrian crossing on a busy one-way street [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 564
18.4 Pedestrian crossing cuts onto sites [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 564
18.5 Traffic signal switching every 10 time periods with singlesimulation run [1]. © IEEE – Reproduced with permission . . . . 565
18.6 Traffic signal switching every 10 time periods averagedover 1000 simulation runs [1]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 566
18.7 Geography of intra-state travel area [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 567
18.8 Plot of status of sites 3–14 against time [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 568
19.1 Behavioral mobility model individual rules andtheir combinations [1]. (a) Pedestrian mobilitymodel: Pedestrians follow a set of three rulesthat express in attractive or repulsive forces.(b) Velocity updating of pedestrians i: Independentacceleration vectors (rule forces) are combined and result
xliv List of Figures
in the acceleration request (net force) used to update thecurrent velocity of i. © IEEE – Reproduced with permission . . . 573
19.2 Traces of pedestrians walking in a corridor with around obstacle [1]. (a) At t=18 s: obstacle avoidance ofpedestrian i and (b) At t=24.5 s: mutual avoidance ofpedestrian i and j. © IEEE – Reproduced with permission . . . . . 578
19.3 Mixed individual and group mobility: Pedestrians meet at arendezvous point (RDV)-point and walk in group formationtoward a common destination but temporarily separatedue to an obstacle [1]. (a) At t = 0 s: initial positions ofpedestrians. (b) At t = 45 s: gathering at RDV-point. (c) Att = 65 s: obstacle avoidance. (d) At t = 100 s: destinationreached. © IEEE – Reproduced with permission . . . . . . . . . . 579
19.4 Evaluation of the realistic pedestrian model versus randomwaypoint approach [1]. (a) Histogram distribution ofvelocity angle deviation in degrees (linear-log scale).The angle deviation is the angle between two successivevelocity vectors taken at 16�t interval where �t = 0.1 s.Each bin accounts for an interval of 2 degrees. (b)Histogram distribution of the link duration of the modifiedRWP compared with the pedestrian model. The radio rangeis set to 5 m. © IEEE – Reproduced with permission . . . . . . . . 581
19.5 Behavior mobility model group simulation algorithm [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 582
20.1 Perfect sampling for restricted random waypoint with onesubdomain A1 [1, 6]. © IEEE – Reproduced with permission . . . 590
20.2 Sampling algorithm for phase φ, previous and next tripendpoints M0 and M1 [1, 6]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
20.3 Random trip using random waypoint mobility pattern [7].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 592
20.4 Node distribution in an example ad hoc network: (a) rescueteam at ground zero and (b) node density distribution [8].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 595
20.5 The pseudo-code of the CMM algorithm [8]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 598
20.6 Node distributions in RWP wrapping (RWP-WRAP)and CMM models: N = 1000, 5 ≤ v ≤ 20 m/s, andTpause = 106 s. α is set to be 1.2 [CMM (1.2)] for (b)and (e) and 1.4 [CMM (1.4)] for (c) and (f). Subfigures(a), (b), and (c) show the layouts before adjustment forsteady state. Subfigures (d), (e), and (f) can be consideredinitial layouts after the steady-state adjustment [8].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 599
List of Figures xlv
20.7 Node connectivity in RWP-WRAP and CMM models:N = 1000, 5 ≤ v ≤ 20 m/s, and Tpause = 106 s [8].(a) After growth, (b) steady state, (c) after growth, (d)steady state. © IEEE – Reproduced with permission . . . . . . . . 602
20.8 PHY capacities in RWP wrapping (RWP-WRAP)and CMM models: N = 1000, 5 ≤ v20 m/s, andTpause = 106 s. The path loss exponent is set to be 2 for (a)and (c) and 3 for (b) and (d). The capture ratio is set to be10 dB for (a) and (b) and 6 dB for (c) and (d) [8]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 603
20.9 PHY link capacities in RWP wrapping (RWP-WRAP)and CMM models: N = 1000, 5 ≤ v ≤ 20 m/s, andTpause = 106 s [8]. (a) Path loss exponent = 2; (b) path lossexponent = 3. © IEEE – Reproduced with permission . . . . . . . 604
20.10 MAC capacities in RWP wrapping (RWP-WRAP)and CMM models: N = 1000, 5 ≤ v ≤ 20 m/s, andTpause = 106 s. The path loss exponent is set to be 2 for (a)and (c) and 3 for (b) and (d). Also, the capture ratio is setto be 10 dB for (a) and (b) and 6 dB for (c) and (d) [8].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 604
20.11 NET capacities in RWP-WRAP and CMM models:N = 1000, 5 ≤ v ≤ 20 m/s, and Tpause = 106 s [8]. ©IEEE – Reproduced with permission . . . . . . . . . . . . . . . . 605
21.1 Graph-based mobility model: (a) City center modeling,(b) Rectangular graph, and (c) Grid graph [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 609
21.2 α values in sample city graph for different transmissionrange [1]. © IEEE – Reproduced with permission . . . . . . . . . 611
21.3 Graph modeling a city center [1]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 614
21.4 Average end-to-end delay for three different routingprotocols [1]. © IEEE – Reproduced with permission . . . . . . . 616
21.5 Packet delivery rates for three different routingprotocols [1]. © IEEE – Reproduced with permission . . . . . . . 618
21.6 Routing protocol packet overhead for three different routingprotocols [1]. © IEEE – Reproduced with permission . . . . . . . 619
21.7 Area graph-based mobility model . . . . . . . . . . . . . . . . . . 62221.8 Results using random waypoint mobility model [2]. (a)
Area size 400 × 400 m. (b) Area size 600 × 600 m.(c) Area size 800 × 800 m. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 627
21.9 Circular area graph [2]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 628
21.10 Results using circular area graph [2]. (a) 200 × 200 m,short waiting time. (b) 300 × 300 m, short waiting time.
xlvi List of Figures
(c) 200 × 200 m, long waiting time. (d) 300 × 300 m, longwaiting time. © IEEE – Reproduced with permission . . . . . . . 629
21.11 Linear area graph [2]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 630
21.12 Delivery ratio using linear area graph [2].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 631
21.13 Results using linear area graph [2]. (a) 200 × 200 m, shortwaiting time. (b) 300 × 300 m, short waiting time. (c) 200× 200 m, long waiting time. (d) 300 × 300 m, long waitingtime. © IEEE – Reproduced with permission . . . . . . . . . . . . 631
22.1 RPGM model’s motion vector of member i of group j . . . . . . . 63922.2 Motion vectors of members of two groups (group 1 and
group 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63922.3 Movement of three mobile nodes using RPGM model . . . . . . . 64022.4 Traveling pattern of one group (three mobile nodes)
using the RPGM model [3]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 641
22.5 Traveling pattern of five groups using the RPGM model[3]. © IEEE – Reproduced with permission . . . . . . . . . . . . 641
22.6 In-place mobility model [1]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 645
22.7 Link-up/down versus mobility [1]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 646
22.8 Clusterhead changing versus mobility [1]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 647
22.9 Throughput of DSDV versus mobility [1]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 648
22.10 Throughput of HSR versus mobility [1]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 648
22.11 Throughput of AODV versus mobility [1]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 649
22.12 Control overhead of DSDV versus mobility [1]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 650
22.13 Control overhead of HSR versus mobility [1]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 650
22.14 Control overhead of AODV versus mobility [1]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 651
22.15 Movements of three mobile nodes using the columnmobility model [3]. © IEEE – Reproduced with permission . . . . 653
22.16 Traveling pattern of mobile nodes using the columnmobility model [3]. © IEEE – Reproduced with permission . . . . 653
22.17 Movements of seven mobile nodes using the nomadiccommunity mobility model [3]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 654
List of Figures xlvii
22.18 Different distances with respect to reference point [14].© ARPN – Reproduced with permission . . . . . . . . . . . . . . 656
22.19 Path traced by node N0 [14]. © ARPN – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 660
22.20 Path traced by node N1 [14]. © ARPN – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 660
22.21 Path traced by node N2 [14]. © ARPN – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 661
22.22 Distance between the reference node and the mobilenodes [14]. © ARPN – Reproduced with permission . . . . . . . . 661
22.23 Snapshot of nodes during third beacon interval [14].© ARPN – Reproduced with permission . . . . . . . . . . . . . . 663
22.24 Snapshot of nodes during 11th beacon interval [14].© ARPN – Reproduced with permission . . . . . . . . . . . . . . 663
22.25 Snapshot of nodes during 31st beacon interval [14].© ARPN – Reproduced with permission . . . . . . . . . . . . . . 663
22.26 Movements of six mobile nodes using the pursue mobilitymodel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 665
22.27 Packet delivery ratio and speed [15]. © IMECS –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 667
22.28 Latency and speed [15]. © IMECS – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 668
22.29 Throughput and speed [15]. © IMECS – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 668
23.1 Mobile nodes represented by their (a) physical coordinatesand (b) velocities [1]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 676
23.2 Network partition and group mobility pattern [1].(a) at t0 = 0 (b) at t1 = 4 (c) at t2 = 8(d) at t3 = 30 (e) at t0 = 0 (marked nodes) (f) att1 = 4 (nodes marked) (g) at t2 = 8 (nodes marked) (h)at t3 = 30 (nodes marked). © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 678
23.3 Guaranteed service availability with partition prediction [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 680
23.4 Perfect accuracy with no misclassification [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 682
23.5 Misclassification [1]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 683
24.1 Initial partition time versus initial distance [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 691
24.2 Partition time versus passing time (D=100 m) [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 691
24.3 Partition time versus passing time (D=1000 m) [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 692
xlviii List of Figures
24.4 Partition time versus passing time (D=2000 m) [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 692
25.1 The structured group mobility model: (a) Node placementand (b) Military units on-the-move . . . . . . . . . . . . . . . . . 697
25.2 Firefighting team in a building: Clearing a room [1].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 700
25.3 Structured groups moving via waypoints, around andthrough obstacles [1]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 702
25.4 Structured groups seeking contact, moving over variousterrain features [1]. © ACM – Reproduced with permission . . . . 703
25.5 Structured groups moving toward an objective, passingdown and then back up a ridge, then over a steep hill [1].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 704
25.6 Link broken by barrier [1]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 705
25.7 Link broken by distance [1]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 706
25.8 Total packets forwarded from any node [1]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 707
25.9 Total packets dropped for any node [1]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 707
25.10 Packets received at destination [1]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 708
25.11 Routing overhead packets [1]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 708
26.1 Virtual track-based group mobility model [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 713
26.2 A partial Los Angeles highway map [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 716
26.3 Abstracted topology from Los Angeles highway map [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 716
26.4 Delivery ratio versus node mobility [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 717
26.5 Throughput versus node mobility [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 717
26.6 End-to-end delay versus node mobility [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 718
26.7 Delivery ration versus number of individual nodes [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 718
26.8 Throughput versus number of individual nodes [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 719
26.9 End-to-end delay versus number of individual nodes [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 719
List of Figures xlix
27.1 drift group mobility model movement patterns:α = 1, μβ = 0.15, σβ = 0.25 (top), andαβ = 0.5 (bottom), Cf = Cl = 2, �t = 1 s [3].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 725
27.2 Warm-up phase: Cumulative distribution function (cdf)of the time needed to get to a target accuracy of 75% forTH=2.5 s [3]. © IEEE – Reproduced with permission . . . . . . . 730
27.3 RG accuracy as a function of the density of independentlymoving users (λ) for W=3, TH = 2.5 2, andTSCAN=2.5 s [3]. © IEEE – Reproduced with permission . . . . . 731
28.1 A comparison of random graph versus scale-free graphs [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 735
28.2 View of the simulated mobility showing spatial distributionof nodes in bounded and boundless situations [1]:(a) Bounded space and (b) Boundless space (wrap-around).© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 739
28.3 Popularity distribution of attractors, in situation ofpopulation growth [1]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 739
29.1 Three distribution force functions [1]: (a) chi-squareddistribution; (b) Rayleigh distribution; (c) F distribution.© SCS – Reproduced with permission . . . . . . . . . . . . . . . 749
29.2 GFMM simulations: (a) NAM illustrating groups of nodesand (b) Zoomed in view of two groups [1]. © SCS –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 753
29.3 GFMM simulations: (a) Mobile node 8 changes directionto avoid collision and (b) Mobile node 8 starts to returnback on path to destination [1]. © SCS – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 753
29.4 GFMM simulations: (a) Mobile nodes 8 and 19 avoidcollision and (b) Mobile node 8 continuing on path withrest of the group [1]. © SCS – Reproduced with permission . . . . 754
29.5 Spatial correlation comparison [1]. © SCS – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 756
29.6 Average link duration [1]. © SCS – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 757
30.1 Comparison of the obtained topology using RWP andRWG model: (a) random waypoint mobility model and(b) random waypoint group mobility model [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 763
30.2 Sequence of three nam screen shots representing evolutionof network topology using SQG mobility model [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 763
l List of Figures
30.3 Performance results for 100-node MANET basic scenariousing different mobility models [1]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 766
30.4 Performance results for 100-node MANET with fivedifferent mobility models as a function of trafficdistribution [1]. © IEEE – Reproduced with permission . . . . . . 767
30.5 Performance results for 100-node MANET with fivedifferent mobility models as a function of maximum nodespeed [1]. © IEEE – Reproduced with permission . . . . . . . . . 768
30.6 Performance results for 100-node MANET with fivedifferent mobility models as a function of number ofgroups [1]. © IEEE – Reproduced with permission . . . . . . . . 769
30.7 Performance results for 100-node MANET with fivedifferent mobility models as a function of simulation areasize [1]. © IEEE – Reproduced with permission . . . . . . . . . . 770
31.1 Tracking of node N1 using mobility information of nodesin a local neighborhood [1]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 782
31.2 Block diagram of mobility tracking algorithm [1].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 783
31.3 Mobile trajectory tracking in mobile ad hoc network [4].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 787
32.1 Autoregressive group mobility model: detection andestimation of AR-1-based individual node mobility andcorrelated node mobility states [1]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 793
32.2 Map of the test site [1]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 798
32.3 Example of group estimation scheme [1]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 799
32.4 Performance of group estimation scheme versus estimationinterval [1]. © IEEE – Reproduced with permission . . . . . . . . 801
32.5 Performance of group estimation scheme versus thresholdrα [1]. © IEEE – Reproduced with permission . . . . . . . . . . . 801
32.6 Performance of correlation index test in terms of probabilityof missed direction [1]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 803
32.7 Performance of correlation index test in terms of probabilityof false alarm [1]. © IEEE – Reproduced with permission . . . . . 803
32.8 Two-tier mobility estimation for node trajectories generatedby RPGM model [1]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 804
33.1 Representation of neighborhood of a unit and flockingrules. (a) Neighborhood of a mobile agent is defined by theangle θ and the distance r. (b) Separation: avoid collision.
List of Figures li
(c) Alignment: steer toward the average heading of flockmates. (d) Cohesion: steer toward average position of flockmates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 810
33.2 Examples of α-lattices and quasi-α-lattices and agent withits neighbors in a spherical neighborhood [3]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 814
33.3 The action and potential functions with finite cut-offs:(a) φα(z) and (b) ψα(z) [3]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 816
33.4 Agent-based representation of obstacles: (a) a wall and(b) a spherical obstacle . . . . . . . . . . . . . . . . . . . . . . . 820
33.5 The in-agent and intra-agent information flow inconstrained flocking: (a) a virtual-leader/followerhierarchical architecture; (b) a peer-to-peer architectureand macro-agent; (c) dynamics; (d) information flowamong species for Algorithm 3; and (e) nested diagrams ofmacro-agents for flocking algorithms in free space and inthe presence of obstacles . . . . . . . . . . . . . . . . . . . . . . 829
33.6 Two-dimensional flocking in free space for n = 150 agents:(a) t = 0 s, (b) t = 0.57 s, (c) t = 2.37 s, (d) t = 3.57 s, (e)t = 4.77 s, and (f) t = 8.97 s [3]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 831
33.7 Two-dimensional flocking in free space for n = 100agents with a dynamic topology: (a) t = 0 s, (b) t = 0.57 s,(c) t = 2.37 s, (d) t = 3.57 s, (e) t = 4.77 s, and (f)t = 8.97 s [3]. © IEEE – Reproduced with permission . . . . . . . 832
33.8 Fragmentation phenomenon in free space for n = 50 agentswith flocking algorithm with no navigational feedback:(a) t = 0 s, (b) t = 0.57 s, (c) t = 1.77 s, (d) t = 2.97 s,(e) t = 8.97 s, and (f) t = 17.97 s [3]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 833
33.9 Snapshots of 3D flocking/automated rendezvous usingflocking algorithm in free space with navigational feedbackfor n = 50 UAVs: (a) t = 0 s, (b) t = 1.17 s, (c) t = 2.37 s,(d) t = 2.97 s, (e) t = 4.17 s, and (f) t = 7.17 s [3].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 834
33.10 Split and rejoin maneuver without drawing the edgesof the (α, β)-net at each moment for n = 150 agents:(a) t = 0 s, (b) t = 5.98 s, (c) t = 15.98 s, (d) t = 22.98 s,(e) t = 30.98 s, and (f) t = 41.98 s [3]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 835
33.11 Topology evolution in split and rejoinmaneuver for n = 150 agents: (a) t = 0 s,(b) t = 5.98 s, (c) t = 15.98 s, (d) t = 22.98 s,
lii List of Figures
(e) t = 30.98 s, and (f) t = 41.98 s [3]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 836
33.12 Squeezing maneuver without drawing the edges of the(α, β)-net at each moment for n = 150 agents: (a) t = 0 s,(b) t = 5.98 s, (c) t = 15.48 s, (d) t = 22.98 s,(e) t = 27.98 s, and (f) t = 43.98 s [3]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 837
33.13 Topology evolution in squeezing maneuver for n = 150agents: (a) t = 0 s, (b) t = 5.98 s, (c) t = 15.48 s,(d) t = 22.98 s, (e) t = 27.98 s, and (f) t = 43.98 s [3].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 837
34.1 Basic characteristics of swarm group mobility model [1]:(a) traces of four nodes in swarm group model and (b) nodespatial distribution. © JUCT – Reproduced with permission . . . . 850
34.2 Inter-nodal distribution and velocity dependency [1]:(a) nodal distribution of inter-node distanceand (b) dependency of the number of nodesin a transmission range and node speed.© JUCT – Reproduced with permission . . . . . . . . . . . . . . 851
35.1 Virtual game-driven mobility model framework andabstraction [1]. (a) virtual game-driven mobility modelframework and (b) abstraction of virtual game-drivenmobility environment. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 859
35.2 Evaluation and select logic used in virtual game-drivenmobility model [1]. © ACM – Reproduced with permission . . . . 861
35.3 Example of two different routing schemes [2]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 867
35.4 Multihop mobile ad hoc network [2]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 868
35.5 Mobility of one user in the plain box scenario with noobstacles [2]. © ACM – Reproduced with permission . . . . . . . 868
35.6 Changes in moving direction [2]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 869
35.7 Available bandwidth of a direct communication betweennode 1 and all other nodes [2]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 869
35.8 Available bandwidth between node 1 and node 13 using asingle-hop connection, multi-hop with shortest path, andmultihop with largest bandwidth [2]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 870
35.9 Multi-hop connectivity c(t) versus time for 13 nodes [2].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 871
35.10 Mean number of hops h(t) versus time for two differentrouting schemes [2]. © ACM – Reproduced with permission . . . 871
List of Figures liii
35.11 Mean multihop bandwidth m(t) versustime for two different routing schemes [2].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 872
35.12 Cumulative stability PDF ψ(t) without any threshold [2].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 873
35.13 Cumulative stability PDF ψ(t) with threshold of12 Mbit/s [2]. © ACM – Reproduced with permission . . . . . . . 873
35.14 Mean stability time versus threshold rθ ,i [2]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 874
36.1 Voronoi diagram and Delaunay triangulation of a setof points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 883
36.2 Inner loop of a kinetic data structure . . . . . . . . . . . . . . . . 88736.3 Separating axis tests for two rectangles in 2D [4]. ©
Chapman and Hall/CRC – Reproduced with permission . . . . . . 89136.4 Human representation of human avatars [5]. © IEEE –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . 89436.5 Overall pipeline of the collision detection algorithm –
different stages performed on the CPU and the graphicsprocessor [5]. © IEEE – Reproduced with permission . . . . . . . 896
36.6 (a) Link i is moving in the reference frame of its parentand the initial and final positions are outlined, (b) offsetof the rule surface, and (c) pipe surfaces [5]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 897
36.7 Collision time estimation [5]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 903
36.8 Environments for performance test of the algorithm [5].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 904
36.9 Benefits of the continuous collision detection algorithmover discrete methods [5]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 904
36.10 Non-recurrent mobility model:(a) contraction, (b) modified contraction, and(c) expansion [10]. © IEEE – Reproduced with permission . . . . 908
37.1 Illustration of TVC mobility model: (a) Three timeperiods and different numbers of communities in eachtime period, (b) two-state Markov chain of local/roamingepoch, (c) Overlapping communities in general case,(d) concentric multiple-tier communities, (e) multiplerandomly placed communities and [1–2]. © IEEE andUniversity of Southern California – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 915
37.2 Spatial distribution of the node (shown as the probabilityfor a node to appear in each 50 × 50 grid block)[1–2]. (a) Randomly placed community, (b) Single-tiercommunity centered at (300, 300) or (700, 700) with one
liv List of Figures
half probability, and (c) Two-tier community centered at(300, 300) or (700, 700) with one half probability. © IEEEand University of Southern California – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 929
37.3 Comparison of theoretical and simulationresults (the average node degree) [1–2].(a) Randomly placed community, (b) Relativeerror for scenarios with fixed single-tier community,and (c) Relative error for scenarios with fixed two-tiercommunities. © IEEE and University of SouthernCalifornia – Reproduced with permission . . . . . . . . . . . . . 930
37.4 Relative error between theoretical and simulation results(the hitting time and meeting time) [1–2]. (a) Hitting time,simple model, (b) Hitting time, multi-tier community,(c) Hitting time, multiple random communnities,(d) Meeting time, simple model, (e) Meeting time,multi-tier community, and (f) Meeting time, multiplerandom communities. © IEEE and University of SouthernCalifornia – Reproduced with permission . . . . . . . . . . . . . 931
37.5 Geographic routing success rate under different mobilityparameter sets and node numbers [1–2]. © IEEE andUniversity of Southern California – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 932
37.6 Message propagation with epidemic routing (the totalpopulation is divided into 2 groups with differentcommunity) [1–2]. © IEEE and University of SouthernCalifornia – Reproduced with permission . . . . . . . . . . . . . 933
37.7 Parameters’ default values [6]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 938
37.8 Prem as a function of n [6]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 939
37.9 Validation of the analytical model (Prem as a functionof n) [6]. © IEEE – Reproduced with permission . . . . . . . . . . 939
37.10 Validation of the analytical model (Prem as a functionof l) [6]. © IEEE – Reproduced with permission . . . . . . . . . . 939
37.11 Prem as a function of wk and w [6]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 940
37.12 Prem as a function of the threshold (uniform distribution ofthe weights) [6]. © IEEE – Reproduced with permission . . . . . . 940
37.13 Prem as a function of wk (uniform distribution of theweights) [6]. © IEEE – Reproduced with permission . . . . . . . . 941
37.14 Prem as a function of wk and f [6]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 942
List of Figures lv
37.15 Prem as a function of f and threshold, in the worst-casescenario (wk = 1−threshold) [6]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 942
37.16 Complementary cumulative distribution function forcontact time [6]. © IEEE – Reproduced with permission . . . . . . 944
37.17 Complementary cumulative distribution function forinter-contact time [6]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 944
37.18 State transition for mobile node’s status in HCBM andCBM [6]. © IEEE – Reproduced with permission . . . . . . . . . 945
37.19 Normalized average time in the IN and OUT states as afunction of q [6]. © IEEE – Reproduced with permission . . . . . . 948
37.20 Markov chain for the standard HCBM [9].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 949
37.21 Simulations versus analysis – standard HCBM [9].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 956
37.22 Simulations versus analysis – modified HCBM [9].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 956
37.23 MLE fitting – 100 communities [9]. © ACM – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 958
37.24 Kernel density estimate with varying number ofcommunities [9]. © ACM – Reproduced with permission . . . . . 959
37.25 Jump size CCDF – varying number of communities [9].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 959
37.26 Kernel density estimate of jump size with rewiring 0.1 [9].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 960
37.27 CCDF for rewiring 0.1 [9]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 960
37.28 CCDF of jump size with varying rewiring [9]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 961
37.29 CCDF of jump size with varying number of nodes [9].© ACM – Reproduced with permission . . . . . . . . . . . . . . . 961
38.1 Orbit-based mobility model scenarios [1]. © StateUniversity of New York – Reproduced with permission . . . . . . 966
38.2 Probabilistic orbit mobility model [1]. © State Universityof New York – Reproduced with permission . . . . . . . . . . . . 971
38.3 States of Markov chain for movements between hubs h andh′ [1]. © State University of New York – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 972
38.4 Steps in preparing a network graph for application ofalgorithm 1 [1]: (a) G = (V, E); (b) GK = (V, Ek), K = 2;(c) Gsp = (V, Esp); (d) G′ = (V, E′) where dotted edgesare special edges: © State University of New York –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 976
lvi List of Figures
38.5 Delivery probability algorithm computation withapproximation [1]. © State University of New York –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 977
38.6 Optimal delivery probability algorithm [1]. © StateUniversity of New York – Reproduced with permission . . . . . . 977
38.7 Examples of analysis of forwarding schemes in staticSOLAR-KSP [1]. © State University of New York –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 979
38.8 Performance of static SOLAR-KSP with varying k [1]:(a) data throughput(%); (b) network byte overhead; and(c) average end-to-end data delay (s). © State University ofNew York – Reproduced with permission . . . . . . . . . . . . . 981
38.9 Protocol performance versus the number of hubs [1]:(a) data throughput(%); (b) network byte overhead; and(c) end-to-end data delay (s). © State University of NewYork – Reproduced with permission . . . . . . . . . . . . . . . . 985
38.10 Protocol performance versus the number of nodes [1]:(a) data throughput(%); (b) network byte overhead; and(c) end-to-end data delay (s). © State University of NewYork – Reproduced with permission . . . . . . . . . . . . . . . . 986
38.11 SOLAR performance versus cache timeout [1]: (a) datathroughput(%); (b) network byte overhead; and (c) end-to-end data delay (s). © State University of New York –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 987
38.12 SOLAR performance versus cache timeout [1]: (a) datathroughput(%); (b) network byte overhead; and (c) end-to-end data delay (s). © State University of New York –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 988
39.1 Power law probability distributions for different values ofq [3]. © ACM – Reproduced with permission . . . . . . . . . . . 997
39.2 Cumulative distribution of packets delivered over the 45days (shaded areas represent days during which packetshave been delivered) [4]. (a) epidemic; (b) opportunistic;(c) random; (d) potato; and (e) MobySpace. © UniversitePierre et Marie Curie – Reproduced with permission . . . . . . . . 1002
39.3 Relative entropy distribution of mobility patterns [4].© Universite Pierre et Marie Curie – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1004
39.4 Community organized by node traces at mobility entropyS = 0, 1, 2, 3 [1]. © AINTEC –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1007
39.5 Nodes organized by contacts at mobility entropyS = 0, 1, 2, 3 [1]. © AINTEC – Reproduced with permission . . . 1007
39.6 Message delivery in PEAR [1]. © AINTEC – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 1010
List of Figures lvii
39.7 Algorithm of message delivery and replica management[1]. © AINTEC – Reproduced with permission . . . . . . . . . . 1011
39.8 Dynamics of potential field construction over the change ofnode-to-node connectivity [1]. © AINTEC – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 1013
39.9 Message delivery rate [1]. © AINTEC – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1015
39.10 Delivery rate at entropy S = 1, 3, 5 [1]. © AINTEC –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1016
39.11 Total message transmission [1]. © AINTEC – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 1017
39.12 Total message transmissions at entropy S = 1, 3, 5 [1].© AINTEC – Reproduced with permission . . . . . . . . . . . . . 1017
39.13 Entropy-based WCA optimized with Tabu search [10].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 1024
39.14 Reaffiliation per unit time versus transmission range,max_disp = 30, number of nodes = 30 [10]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1026
39.15 Reaffiliation per unit time versus maximum displacement,number of nodes = 30, tx_range=30 [10]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1027
39.16 Convergence graph of TS and SA on a problem instance[10]. © IEEE – Reproduced with permission . . . . . . . . . . . . 1028
39.17 Average number of cluster versus maximum displacement,number of nodes = 30, and tx_range=30 [10]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1028
39.18 Entropy-based different priority controllablearchitecture [2]. © Springer-Verlag Berlin Heidelberg –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1030
39.19 Entropy-based fuzzy routing in MANET [2]. ©Springer-Verlag Berlin Heidelberg – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . 1031
39.20 Data transmission rate versus mobile node’s speed [2].© Springer-Verlag Berlin Heidelberg – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1031
39.21 Average end-to-end delay versus mobile node’s speed [2].© Springer-Verlag Berlin Heidelberg – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1031
40.1 One disconnected route from D to F [5]. (a) from timet0 to t1, (b) from time t1 to t2, and (c) from time t2 to t3.© Washington University – Reproduced with permission . . . . . 1043
40.2 Proactive service provision through relocation (TR =tr, TS = te, μHost A = motion profile of Host A, μHost B =motion profile of Host B, and μuser =
lviii List of Figures
motion profile of user) [5]. © Washington University –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1044
40.3 Continuous connectivity through migration (TS = ts,TE = te, TM = tm, μHost A = motion profile of Host A,μHost B = motion profile of Host B, andμuser = motion profile of user) [5].© Washington University – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1045
40.4 SPAWN architecture with knowledge managementinfrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1047
40.5 KnowledgeDisseminator and KnowledgeAggregatorarchitecture [3]. © ACM – Reproduced with permission . . . . . . 1049
40.6 Knowledge management system architecture [3]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 1049
40.7 Interaction of various components of knowledgemanagement system [3]. © ACM – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1052
40.8 Service discovery architecture view [3]. © ACM –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1054
40.9 Effect of communication range on knowledge base sizeand percentage of hosts satisfied [5]. © WashingtonUniversity – Reproduced with permission . . . . . . . . . . . . . 1058
40.10 Effect of host density on knowledge base size andpercentage of hosts satisfied [5]. © Washington University– Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1059
40.11 Number of meetings, knowledge base size, and satisfiedhosts (random walk model) [5]. © Washington University –Reproduced with permission . . . . . . . . . . . . . . . . . . . . 1060
40.12 Number of meetings, knowledge base size, and satisfiedhosts (random waypoint model) [5]. © WashingtonUniversity – Reproduced with permission . . . . . . . . . . . . . 1060
40.13 Effect of requirement length on percentage of hosts beingsatisfied [5]. © Washington University – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1061
40.14 Effect of future window size on percentage of hosts beingsatisfied [5]. © Washington University – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1062
List of Tables
3.1 Notations used in random walk mobility model . . . . . . . . . . . 383.2 Conditional density probabilities for the random walk
mobility model [12] . . . . . . . . . . . . . . . . . . . . . . . . . 423.3 Basis functions for X- and Y-coordinates [12] . . . . . . . . . . . . 444.1 Notations for random waypoint mobility model . . . . . . . . . . . 684.2 Mapping between pause probability and expected pause time
(square area) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 814.3 Notations for the generic mobility model: RWP and
variations of RWP . . . . . . . . . . . . . . . . . . . . . . . . . . 924.4 Polynomial approximations of the spatial RWP node
distribution in unit disk [19]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.1 Notation for smooth random mobility model . . . . . . . . . . . . 1265.2 Example parameter for speed control [6]. © ACM –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 1285.3 Example parameters for direction parameter [6].
© ACM – Reproduced with permission . . . . . . . . . . . . . . . 1316.1 Notations used in geographic constraint mobility models . . . . . . 1686.2 Parameters setting of mobility models [5]. © IEEE –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 1786.3 Mobility parameters settings for simulation experiments [5].
© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 1826.4 Power attenuation value [3]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1936.5 Number of failed connections [3]. © IEEE – Reproduced
with permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1986.6 Connection threshold for node 1 [10]. © Academic
Publisher – Reproduced with permission . . . . . . . . . . . . . . . 2056.7 Connection threshold for node 2 [10]. © Academic
Publisher – Reproduced with permission . . . . . . . . . . . . . . . 2066.8 Sample values of pause time (PT) tp [10]. © Academic
Publisher – Reproduced with permission . . . . . . . . . . . . . . . 2087.1 Notations used for random mobility direction model . . . . . . . . 225
lix
lx List of Tables
8.1 Notations used in deterministic mobility model . . . . . . . . . . . 2469.1 Notations used in PDM model . . . . . . . . . . . . . . . . . . . . 2669.2 Simulation parameters [2]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2849.3 Prediction result of next cell crossing [2]. © IEEE –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 2859.4 Global prediction result . . . . . . . . . . . . . . . . . . . . . . . 28510.1 Notations used in RGM mobility model . . . . . . . . . . . . . . . 31211.1 Notations for semi-Markov smooth mobility model . . . . . . . . . 34612.1 Notations used in BSA mobility model . . . . . . . . . . . . . . . 38012.2 Simulation results for λLCR.TLD/2-nominal average node
degree are set to ρπr2 = 6.0 [4]. © IEEE – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400
13.1 Notations used in fluid-flow mobility . . . . . . . . . . . . . . . . 40513.2 SOR algorithm [6]. © Inderscience – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41814.1 Notations used in gravity mobility model . . . . . . . . . . . . . . 44314.2 Limits of fluid law [6]. © ACM – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45915.1 Notations used in vector mobility model . . . . . . . . . . . . . . . 48316.1 Notations used in correlated diffusion mobility model . . . . . . . . 49617.1 Notations used in particle-based mobility model . . . . . . . . . . . 52717.2 Simulation parameter values [6]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54817.3 Efficiency of data transmission [6]. © IEEE – Reproduced
with permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55418.1 Notations used in hierarchical influence mobility model . . . . . . . 55819.1 Notations used in behavioral mobility model . . . . . . . . . . . . 57220.1 Notations used in steady-state generic mobility model . . . . . . . 58521.1 Notations used in area and graph-based mobility model . . . . . . . 60821.2 Length of sample graphs, for gross area
1250 × 900 m2 [1]. © IEEE – Reproduced with permission . . . . 61021.3 DSDV simulation parameters [1]. © IEEE – Reproduced
with permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61221.4 DSR simulation parameters [1]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61321.5 AODV simulation parameters [1]. © IEEE – Reproduced
with permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61321.6 Simulation parameters [1]. © IEEE – Reproduced with
permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61521.7 Simulation parameters for all experiments [2]. © IEEE –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 62621.8 Network density using random waypoint mobility model [2].
© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 626
List of Tables lxi
21.9 Density of sub-scenarios of scenario 1 – circle [2]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 629
21.10 Density of sub-scenarios – line [2]. © IEEE – Reproducedwith permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . 630
22.1 Notations used in RPGM model . . . . . . . . . . . . . . . . . . . 63822.2 Groups specified in the RPGM model [3]. © IEEE –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 64222.3 Distance between pair of nodes in each beacon interval [14].
© ARPN – Reproduced with permission . . . . . . . . . . . . . . . 66222.4 Parameters used for simulation scenarios [15]. © IMECS –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 66723.1 Notations used in RVMG model . . . . . . . . . . . . . . . . . . . 67223.2 Sequential clustering algorithm [1]. © IEEE – Reproduced
with permission . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68124.1 Notations used in RVAM model . . . . . . . . . . . . . . . . . . . 68625.1 Notations used in the structured group mobility model . . . . . . . 69627.1 Notations used in the drift group mobility model . . . . . . . . . . 72227.2 Average lapse of time and standard deviation to get to a
RG – accuracy of x% for TH = 2.5 s, and TSCAN= 5 s [3].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 729
27.3 Average lapse of time and standard deviation to get to aRG – accuracy of x% for W=3, and TSCAN = 2.5 s [3].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 730
28.1 Notations used in the gathering group mobility model . . . . . . . . 73429.1 Notations used in group force mobility model . . . . . . . . . . . . 74531.1 Notations used in autoregressive mobility model . . . . . . . . . . 77631.2 RMSE of mobile trajectory over 500 sample experiments
[4]. © IEEE – Reproduced with permission . . . . . . . . . . . . . 78832.1 Notations used in the autoregressive group mobility model . . . . . 79232.2 Direction and estimation of groups [1]. © IEEE –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 79932.3 RMSE of group estimation algorithm [1]. © IEEE –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 80433.1 Notations used in flocking mobility models . . . . . . . . . . . . . 81134.1 Notations used in swarm group mobility model . . . . . . . . . . . 84435.1 Nations used in virtual game-driven mobility models . . . . . . . . 85835.2 Parameters for IEEE 802.11a simulator [2]. © ACM –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 86736.1 Notations used in non-recurrent mobility models . . . . . . . . . . 87936.2 Types of bounding volume hierarchies [4]. © Chapman and
Hall/CRC – Reproduced with permission . . . . . . . . . . . . . . 88936.3 Average performance of the various steps of the algorithm
for a 16-link avatar moving in the virtual environmentcomposed of hundreds of thousands of polygons [5].© IEEE – Reproduced with permission . . . . . . . . . . . . . . . 905
lxii List of Tables
37.1 Notations used in time-variant community mobility model . . . . . 91337.2 Parameters for the scenarios in the simulation [1] . . . . . . . . . . 92838.1 Notations used in orbit-based mobility model . . . . . . . . . . . . 96638.2 Mobility parameters for orbit-based mobility [1]. © State
University of New York – Reproduced with permission . . . . . . . 96838.3 Simulator parameters [1]. © State University of New York –
Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 98139.1 Notations used in entropy-based mobility model . . . . . . . . . . . 99439.2 Simulation parameters [4]. © Universite Pierre et Marie
Curie – Reproduced with permission . . . . . . . . . . . . . . . . . 100039.3 Results with randomly sampled users [4]. © Universite
Pierre et Marie Curie – Reproduced with permission . . . . . . . . 100139.4 Results with the most active users [4]. © Universite Pierre et
Marie Curie – Reproduced with permission . . . . . . . . . . . . . 100339.5 Results with users having different entropies [4].
© Universite Pierre et Marie Curie – Reproduced withpermission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1004
39.6 Results with space reduction – l is the number of mostsignificant components taken into account [4]. © UniversitePierre et Marie Curie – Reproduced with permission . . . . . . . . 1005
39.7 Mean values of simulation results, max_disp =30, max_degree = 10, and tx_range = 30 [10]. © IEEE –Reproduced with permission . . . . . . . . . . . . . . . . . . . . . 1029
39.8 Entropy-based fuzzy scheduler rules [2]. © Springer-VerlagBerlin Heidelberg – Reproduced with permission . . . . . . . . . . 1030
40.1 Notations used in KDM model . . . . . . . . . . . . . . . . . . . . 1036
About the Author
Radhika Ranjan Roy is an electronics engineer, Communications and ElectronicsResearch and Development Center (CERDEC), Fort Monmouth, NJ, USA, since2009. He is leading his research and development efforts in the mobile ad hocnetworks and supporting array of Army’s Nationwide and Worldwide WarfighterNetworking Architectures and participating in technical standards development inmultimedia/real-time services collaboration, IPv6, radio communications, enter-prise services management, and information transfer of Department of Defense(DoD) technical working groups (TWGs). He received his PhD in electrical engi-neering with major in computer communications from the City University of NewYork, NY, USA, in 1984 and MS in electrical engineering for the NortheasternUniversity, Boston, MA, USA, in 1978. He received his BS in electrical engineeringfrom the Bangladesh University of Engineering & Technology, Dhaka, Bangladesh,in 1967.Prior to joining CERDEC, Dr. Roy worked as the lead system engineer at CACI,Eatontown, NJ, from 2007 to 2009 and developed Army Technical ReferenceModel (TRM), Army Enterprise Architecture (AEA), DoD Architecture Framework(DoDAF), and Army LandWarNet (LWN) Capability Sets, and technical stan-dards for Joint Tactical Radio System (JTRS), Mobile IPv6, MANET, and SessionInitiation Protocol (SIP) supporting Army Chief Information Officer (CIO)/G-6. Dr.Roy worked as senior system engineer, SAIC, Abingdon, MD, from 2004 to 2007supporting modeling, simulations, architectures, and system engineering of manyarmy projects: WIN-T, FCS, and JNN.During his career, Dr. Roy worked in AT&T/Bell Laboratories, Middletown, NJ,as senior consultant from 1990 to 2004 and led a team of engineers in designingAT&T’s worldwide VoIP/multimedia communications network architecture con-sisting of wired and wireless from preparation of request for Information (RFI),evaluation of vendor RFI responses and interactions with all selected major ven-dors related to their products. He participated and contributed in the developmentof VoIP/H.323/SIP multimedia standards in ITU-T, IETF, ATM, and Frame Relaystandard organizations.Dr. Roy worked as a senior principal engineer in CSC, Falls Church, VA, from1984 to 1990 and worked in design and performance analysis of the US Treasury
lxiii
lxiv About the Author
nationwide X.25 packet switching network. In addition, he designed many net-work architectures of many proposed US Government and Commercial Worldwideand Nationwide Networks: Department of State Telecommunications Network(DOSTN), US Secret Service Satellite Network, Veteran Communications Network,and Ford Company’s Dealership Network. Prior to CSC, he worked from 1967 to1977 as deputy director, Design, in PDP, Dhaka, Bangladesh.Dr. Roy’s research interests include mobile ad hoc networks, multimedia communi-cations, peer-to-peer networking, and quality-of-service. He has published over 50technical papers and is holding or pending over 30 patents. He also participates inmany IETF working groups. He lives in historical district of Howell Township, NJ,with his wife Jharna.