TutorialsSunday and Monday, September 14 and 15, at MobiCom 2003 will feature a number of full-day and half-day tutorials on the latest research areas and background topics in mobile computing and networking. The following tutorials will be presented at MobiCom 2003:
- Sunday, September 14 -- Full-day tutorial
(8:30 AM-5:00 PM):
- Monday, September 15 -- Morning half-day tutorials
(8:30 AM-12:00 PM):
- Monday, September 15 -- Afternoon half-day tutorials
(1:30 PM-5:00 PM):
|Tutorial 1 -- Sunday, September 14, 8:30 AM-5:00 PM|
Secure Routing in Ad Hoc Networks
Carnegie Mellon University, USA
Rice University, USA
A number of routing protocols have been proposed for routing in ad hoc networks; however, until recently, security in such networks has not received much attention from the research community. In this tutorial, we provide an overview of security and ad hoc networks, and security threats applicable to ad hoc networks. We then cover a number of solutions to these threats in detail, including solutions to the problems of key setup, secure network routing, intrusion detection, and solutions to some advanced attacks. We also will discuss the remaining challenges in the area.
- Introduction to Security
- Basic Cryptographic primitives
- Introduction to Ad Hoc Networks
- Attacker Model
- Key Setup
- Ariadne-style KDC
- Secret Sharing: Zhou & Haas, Lu et al
- Resurrecting Duckling
- Secure Ad Hoc Network Routing
- Awerbuch et al's
- Intrusion Detection
- Mitigating Routing Misbehavior
- Advanced attacks
- Same Metric Fraud
- Defense mechanisms
- RAP (Rushing Attack Prevention)
- Super SEAD
- Other Issues
- MAC-layer security
Intended Audience:This tutorial will appeal to a large range of researchers and practitioners of the mobile networking community who are interested in security and ad hoc network routing protocols. For practitioners, this tutorial will provide guidelines and many examples on how to design secure ad hoc network routing protocols. For researchers, this tutorial will review the current state-of-the-art in secure ad hoc network routing, and will outline future research directions. No prior knowledge of ad hoc networking or security and cryptography is assumed, but some familiarity with protocols and mobile networking is recommended.
Speaker Bios:Adrian Perrig is an Assistant Professor in Electrical and Computer Engineering, Engineering and Public Policy, and Computer Science at Carnegie Mellon University. He earned his PhD in Computer Science from Carnegie Mellon University, and spent three years during his PhD at University of California at Berkeley. He received his BS in Computer Engineering from the Swiss Federal Institute of Technology in Lausanne (EPFL). Adrian's research interests revolve around building secure systems and include network security, security for sensor networks, and mobile applications. More information about his research is available at http://www.ece.cmu.edu/~adrian.
Yih-Chun Hu received his B.S. from the University of Washington in 1997 and his Ph.D. from Carnegie Mellon University in 2003. In his thesis work at Carnegie Mellon, he focused on security and performance in wireless ad hoc networks. Yih-Chun's research interests include mobility and security in wired and wireless networks. He is currently a postdoctoral research fellow at Rice University and is a member of the ACM.
|Tutorial 2 -- Monday, September 15, 8:30 AM-12:00 PM|
Public-Area Wireless Networks:
Security, Deployment, Performance Optimization, and Pricing Issues
University of California, San Diego, USA
The vision of pervasive ubiquitous computing, where users have network access anytime, anywhere, is being enabled by deployments of high-speed wireless networks in common places of congregation such as airports, malls, hotels, parks, and arenas. Such Public-Area Wireless Networks (PAWNs) have recently become ubiquitously deployed both through open community efforts (such as NYCwireless and SeattleWireless) and through commercial, business-driven wireless ISPs (such as Cometa and Wayport). Nevertheless, Wi-Fi as a technology has to overcome a number of technological and economic hurdles before PAWNs become an attractive and complementary networking solution. We outline these challenges below.
Interoperability: Since wireless LANs operate in the unlicensed radio spectrum where many different standards like IEEE 802.11b, a, and g exist, future wireless networking equipment will have to support inter-operation between all these standards in their transmitters and receivers. Equipment vendors are partially addressing this challenge by integrating support for newer standards in their equipment, while allowing backward compatibility with current standards.
Security and Authentication: Despite some well-known solutions for 802.11 security through per-packet authentication, several attacks exist which leave PAWNs vulnerable to malicious users. Recently, researchers have discovered several means to launch DoS attacks on 802.11 systems, thus exposing their vulnerability. Therefore, PAWN administrators need an understanding of the various cryptographic and security issues relevant to 802.11-based networks.
Network Performance Optimization: Although Wi-Fi has been considered as a high-speed networking technology, huge differences exist between the theoretical range and capacity of these networks and their respective performance obtained in practice. The dynamically varying nature of the wireless channel and attenuation of RF signals due to obstructions greatly reduce the range and capacity offered by the access points. As the use of these networks spreads beyond simple data transfer to data- and performance-intensive multimedia applications, the need to address capacity planning issues will be increasingly important. Furthermore, the available network capacity will need to take into account dynamically-varying and location-dependent user load, which might impact overall network utilization.
PAWN Business Models: There is a growing need for a viable business model that can generate revenue and hence encourage businesses and organizations to install PAWNs on their premises. To address this, PAWN providers will have to introduce subscription-based charging for customers and incorporate policies that include security, performance, and quality of service into the pricing model. Also, since no single nationwide Wi-Fi service exists, a PAWN business service subscription model will have to include roaming agreements between providers.
While all of the above issues are being addressed in part by industry efforts, the research community as a whole has not been exposed to them. This tutorial will span all of the above aspects of designing and deploying such PAWNs and will cover research solutions to key challenges that enable their widespread deployment: security, authentication, network capacity planning, pricing, and bandwidth provisioning.
- Introduction to PAWNs
- Differences with cellular systems
- Differences between PAWNs and Wireless MANs
- Ubiquitous Deployments and War Chalking
- PAWN Architectures and Goals
- Wi-Fi Spectrum and Standards
- Open vs. Secure Access
- Monitoring and Enforcement
- Security, Authentication, Authorization
- Layer-2 Approaches, WEP, 802.1X
- Layer-3 Approaches, IPSec/IKE
- EAP, RADIUS and 802.11i Proposals
- Possible Attacks on 802.11
- Capacity Planning
- Industry Solutions
- PAWN Workload Characterizations
- 802.11 Throughput Characterization
- Load Balancing
- Roaming and Handoffs in PAWNs
- User Bandwidth Allocation
- Formal Models
- Service Models and Pricing
- Commercial Solutions
- Pricing Studies
- IETF Standardization Efforts
Intended Audience:The tutorial should be of interest to both network designers and end system implementers in academia and industry. It will also benefit all researchers interested in the areas related to local- and wide-area wireless communications and mobile computing.
Speaker Bio:Anand Balachandran is a graduating Ph.D. student at the University of California, San Diego, and will be beginning a position of Ubiquitous Computing Researcher at Intel Research, Seattle, in October 2003. His doctoral research has addressed issues concerning the development and deployment of Public-Area Wireless Networks through experimental study and the implementation of novel algorithms. His research interests include wireless networking systems; wireless Internet; infrastructure and ad hoc networks; and mobile computing. Anand received his Bachelor of Technology from the Indian Institute of Technology, Madras in 1995 and his Master of Science degree from Columbia University in 1997. He is a member of the IEEE Communications Society and a member of the ACM. For more information, visit http://www-cse.ucsd.edu/users/abalacha/.
|Tutorial 3 -- Monday, September 15, 1:30 PM-5:00 PM|
IP Mobility Support in a Mobile InternetRajeev Koodli
Nokia Research Center, Mountain View, CA, USA
The convergence of wireless communication and the Internet is leading the way for a Mobile Internet, which brings two important benefits to its users; un-tethered communication and mobility. While the convenience of un-tethered communication may be obvious, providing support for mobility is subtler. From a user's perspective, mobility translates to an ability to be reachable in spite of movement across networks, as well as an ability to maintain existing communication during such movements. Hence, mobility solution consists of support for roaming, which provides "always-on" global reachability, and support for handover, which provides existing session continuity.
In this tutorial, we will consider network layer support for roaming and handover. We recognize that the root of the mobility problem is the need to change an IP address as a user device roams from a network to another. The 3G networks evolved from GPRS solve this problem by avoiding IP address change; the Mobile IP protocol from the IETF solves the problem by addressing the outcome of IP address change. We will investigate the relative merits of these two mobility models. We will show how network-managed mobility is best suited for single access networks such as GPRS, and host-managed mobility is best applicable in multi-access networks, such as GPRS and Wireless LANs. However, both are deficient in supporting real-time applications; the tutorial will go through advanced handover topics such as "Fast Handovers," which minimizes latencies associated with connectivity establishment and location update, and "Context Transfers," which smoothes handover effect on transport protocol performance.
- Introduction to IP Mobility
- issues to consider
- problem introduction
- connectivity establishment
- location update
- scope of rest of the tutorial
- "Link-layer" solution: GPRS
- basic architecture
- connectivity establishment
- access link connectivity, authentication
- IP connectivity: "PDP Context" establishment
- different handover scenarios and location updates
- intra-SGSN handover
- inter-SGSN handover
- Network-layer solution: Mobile IP
- basic principles, terminology
- connectivity establishment
- movement detection
- IP address configuration
- access control considerations
- location update
- new IP address "registration"
- security considerations
- packet forwarding
- route optimization
- optimizing data path between a mobile and its correspondent(s)
- security issues
- "Return Routability" principle
- Advanced Handover Topics
- fast handovers
- reducing connectivity establishment latency
- removing the effect of location update latency
- performance considerations in WLAN
- context transfers
- smoothing the effect of handover by relocating network-resident state
- synchronizing with fast handovers
- example feature context: header compression
- hierarchical Mobile IP (time-permitting)
- reducing the number of signaling messages
- location privacy
- fast handovers
- Overall summary
Intended Audience:Researchers and engineers interested in understanding the principles of the IP Mobility architecture, its components, and recent developments. No prior understanding of the field is necessary, although a basic knowledge of Internet Protocol networking is needed.
Speaker Bio:Rajeev Koodli is a Principal Scientist at Nokia Research Center, Mountain View, CA, USA. His current research interests are in IPv6 mobile networking and ad hoc networking. Apart from the design of Mobile Internet architecture, his focus areas include header compression, QoS, service discovery, and IP performance characterization. He is the editor of the IETF draft "Fast Handovers for Mobile IPv6" and co-author of the IETF documents "Context Transfer Protocol" and "Loss Pattern Sample Metrics" (RFC 3357). He has published many journal and conference publications and (co)authored numerous patent applications. He was an invited member of IEEE Wireless Delegation to China in 2000, a visiting researcher at AT&T Bell Labs, and a recipient of special honor for receiving graduate research grant from Bell Laboratories. He has MS and Ph.D. degrees from the University of Massachusetts, Amherst.
|Tutorial 4 -- Monday, September 15, 8:30 AM-12:00 PM|
Mobile Ad Hoc Networking
|Elizabeth M. Belding-Royer
University of California, Santa Barbara, USA
Hewlett-Packard Laboratories, USA
What makes ad hoc networks different from wired and cellular networks, and how can network communication successfully be accomplished in these networks? What characteristics of ad hoc protocols result in their varied performances? These are some of the questions we will address in this tutorial on ad hoc networking. This tutorial will begin with a description of the characteristics and applications of wireless ad hoc networks that distinguish them from their wired counterparts and from cellular networks. We will then cover in depth many well known protocols, including MAC layer, routing, and multicast protocols. The problems of TCP in ad hoc wireless networks, and some proposed solutions, will also be highlighted. Finally, we will conclude the tutorial by discussing current challenges to ad hoc networking, as well as IETF standardization efforts.
- overview, challenges, and applications of ad hoc networking
- Medium Access Control Protocols
- CSMA, MACA, IEEE 802.11
- Unicast Routing
- Proactive routing
- On-demand routing
- Location-based routing
- Cluster/Hierarchical routing
- Multicast Routing
- Tree-based protocols
- Mesh-based protocols
- Location-based protocols
- Transport Layer Issues
- TCP performance in ad hoc networks
- Techniques to improve TCP performance
- Other Issues
- Energy/Power conservation
- Security for ad hoc networks
- Implementation efforts and testing issues
- Future Direction and Open Problems
- Asymmetric routing and unidirectional links
- IETF MANET standardization efforts
- Lots of others...
Intended Audience:This tutorial is intended for researchers and engineers in both industry and academia, and for anyone who would like a deeper understanding of mobile networking and the current state of research in this area. The tutorial is designed to provide an overview of the issues related to mobile ad hoc networking, as well as in-depth coverage of current efforts in enabling communication, at the MAC, network, and transport layers, in these networks.
Speaker Bios:Elizabeth M. Belding-Royer is an Assistant Professor in the Department of Computer Science at the University of California, Santa Barbara. She completed her Ph.D. in Electrical and Computer Engineering at UC Santa Barbara in 2000. Elizabeth's research focuses on mobile networking, specifically routing protocols, security, scalability, and adaptability. Elizabeth is the author of numerous papers related to ad hoc networking and has served on many program committees for networking conferences. Elizabeth is currently the co-chair of the IRTF Ad hoc Network Scalability (ANS) Research Group and is also on the editorial board for the Elsevier Science Ad hoc Networks Journal. She is also the recipient of a 2002 Technology Review 100 award, awarded to the world's top young investigators.
Sung-Ju Lee is a research scientist/engineer at the Mobile & Media Systems Lab (MMSL) of Hewlett-Packard Laboratories. S.J. received his Ph.D. in Computer Science from University of California, Los Angeles. His dissertation focused on mobile ad hoc wireless networks. S.J. published over thirty papers in the field of mobile networks. He is a co-guest editor of the Wireless Communications and Mobile Computing's special issue on mobile ad hoc networking, is an associate editor on ad hoc networks for ACM SIGMOBILE Mobile Computing and Communications Review (MC2R), is an associate editor for Elsevier Science's Ad Hoc Networks Journal, is a TPC co-chair for the first ACM workshop on Wireless Mobile Applications and Services on WLAN Hotspots (held in conjunction with MobiCom 2003), and serves as a technical program committee and organizing committee member of various networking conferences. His research interests include mobile networking and computing, ad hoc networks, media distribution networks, overlay networks, and public WLAN.
|Tutorial 5 -- Monday, September 15, 1:30 PM-5:00 PM|
Topology Control in Wireless Ad Hoc NetworksPaolo Santi
Carnegie Mellon University, USA (on leave from IIT-CNR, Italy)
The topology control problem in wireless ad hoc networks is to choose the transmit power of each node in such a way that energy consumption is reduced and some property of the communication graph (typically, connectivity) is maintained. Besides reducing energy consumption, topology control increases the capacity of the network, due to reduced contention to access the wireless channel. In fact, it is known that it is better, from the network capacity point of view, to send packets along several short hops rather than using long hops. Given the limited availability of both energy and capacity in ad hoc networks, and especially in wireless sensor networks, topology control is thus considered as a major building block of forthcoming wireless networks.
This tutorial has three major goals. First, we will give the attendees an organic view of the considerable body of literature devoted to topology control in ad hoc networks. Second, the tutorial will provide the attendees with a "topology control toolkit," i.e., with the expertise needed in the design and analysis of effective topology control mechanisms. The theoretical tools that compose this toolkit are derived from the theories of applied probability, computational geometry, and distributed systems; furthermore, we will discuss in details the requirements that any topology control protocol should meet in order to be implementable in a real ad hoc network. Finally, we will outline the limitations of current approaches, paying particular attention to those factors that hamper the application of topology control techniques in a realistic scenario. This way, we will disclose several directions of further research in the field.
- Motivation: the need for topology control in wireless ad hoc networks
- A simplified but widely accepted ad hoc network model
- Topology control: a taxonomy
- Stationary networks
- The critical transmitting range for connectivity
- Energy efficient communications: building spanners for energy-efficient unicast/multicast/broadcast
- Topology control protocols: "ideal" properties of a TC protocol and state-of-the-art solutions
- Mobile networks
- Mobility models
- The mobile critical transmitting range
- Topology control with mobility
- Open issues
- More realistic network and energy models
- More accurate analysis of mobile networks: is mobility beneficial or detrimental for topology control?
- Considering the effect of multi-hop data traffic
- Towards an implementation of topology control: level-based approaches
- per-packet topology control: the COMPOW and CLUSTERPOW protocols
- periodic topology control: the KNeighLev protocol
Intended Audience:The tutorial will address both theoretical and technological aspects related to topology control. For this reason, this tutorial should appeal to researchers with either a computer science or engineering background, as well as to people from the industrial community. Graduate and undergraduate students, and any professional who is interested in entering into this fundamental field of ad hoc networking research, are especially welcome.
Speaker Bio:Paolo Santi received the MS degree and the PhD in Computer Science from the University of Pisa, Italy, in 1994 and 2000, respectively. During the PhD, his research interest focused on fault-tolerant computing. Starting from the end of 2000, Dr. Santi's interests shifted to wireless ad hoc networks, first from a dependable computing point of view. The interest in wireless ad hoc networks grew wider during the six months (January-June 2001) visit at the School of Electrical and Computer Engineering, Georgia Institute of Technology. On Sept. 2001, Dr. Santi joined the Italian National Research Council in Pisa. Since then, his research interests broadened, including also more theoretical topics, such as combinatorial auctions. Currently, Dr. Santi is visiting the Department of Computer Science at Carnegie Mellon University. The properties of wireless ad hoc networks considered or currently under investigation include topology control, network lifetime evaluation, cooperative strategies for reducing energy consumption, game theoretic aspects of ad hoc networks, and the node spatial distribution of mobile networks. Although Dr. Santi's interest in wireless ad hoc networks is relatively recent, he already contributed several papers in major journals and conferences in the field. Dr. Santi is a member of ACM and SIGMOBILE.