MobiCom 2005 will include demonstrations of some of the best state-of-the-art research in the field of mobile computing and wireless and mobile networking. The following demos have been accepted for presentation at the conference:

• S-MobiEmu - Evaluating Secure Systems for Mobile Ad Hoc Networks
  Y. Zhang (HRL Laboratories, LLC), Y. Huang (Georgia Tech), W. Lee (Georgia Tech)

  Developing and evaluating secure MANET (mobile ad-hoc networks) in real systems is a complex process that involves careful design of attack test cases and security countermeasures. In this demo, we demonstrate a software framework in the secure MANET routing domain that can be used as a development environment to automate this process. In particular, we demonstrate an extensible attack library that includes a full set of basic attacks at its core and show that it is possible to compose compound attacks from the basic elements in a systematic way. We use the development of an Intrusion Detection System (IDS) as a case study. Our successful experience confirms that the platform can greatly facilitate the development of security solutions on MANET.

• Inter-Vehicle Communication - Improving Driver Control and Vehicle Safety
  A. Festag (NEC Europe Ltd.), R. Schmitz (NEC Europe Ltd.), R. Baldessari (NEC Europe Ltd.)

  Inter-vehicle communication based on wireless communication shows a strong potential to enhance both safety and comfort of road users. Existing approaches for routing in mobile ad hoc networks, like AODV and DSR, can not cope with the high dynamics of vehicular environments. We demonstrate the benefits of position-based routing (PBR) in vehicular ad hoc networks (VANETs) in an experimental testbed with emulated mobility of nodes. We demonstrate a) the capability of PBR to react quickly to route changes, b) the'geocasting' of data to nodes in geographical areas in combination with store & forward to improve reliability in scenarios with a low density of cars, and c) the benefits of traffic differentiation and prioritization of data of data from 'safety of life' applications.

• Autoconfiguration and Distributed Firewalling for MANET
  F. Frosali (Business Applications & Services Telecomitalia LAB), G. Freguglia (Business Applications & Services Telecomitalia LAB), L. Costa (Business Applications & Services Telecomitalia LAB), S. Ruffino (Wireless Networking Telecomitalia LAB), P. Stupar (Wireless Networking Telecomitalia LAB)

  MANET GAA-DFA is an integrated solution for automatic configuration of network addresses and distributed traffic filtering suited for MANETs connected by means of gateways to an infrastructure network (e.g. the Internet). It comprises two mechanisms: MANET-GAA, a method for IPv6 global address configuration in a MANET endowed with multiple gateways, and MANET-DFA, a firewall solution based on a client-server distributed architecture. The first one interacts with routing protocol providing global prefixes to MANET nodes and minimizing the latency experienced by a node after it changes its global address. The second one dynamically configures profile-based filtering rules on the MANET nodes which apply only necessary policy based on address and topology changes information provided by routing protocol.

• Visualizing Wireless ns-2 Traces in 3D
  B. Scheuermann (University of Mannheim), H. Fuessler (University of Mannheim), M. Transier (University of Mannheim), M. Mauve (University of Duesseldorf), W. Effelsberg (University of Mannheim)

  Network simulation is an important research tool for Mobile Ad-Hoc Networking. For users of this tool, visualization techniques are useful to help to understand the events occurring in the non-real-time simulator. In this demo, we present Huginn, a powerful and easily customizable visualizer for ns-2 traces. Huginn uses three dimensions to show visual representations of the mobile nodes and their communication events.
  
  The demo will consist of a running version of Huginn and a prepared set of ns-2 traces ready for visualization. We will also show the configurability of Huginn by means of prepared or visitor-driven examples.

• The Illionois GRACE Project: Energy-efficient Cross-layer System Design
  A.F. Harris, W. Jeon, D.G. Sachs, V. Vardhan, S.V. Adve, D. Jones, R. H. Kravets, K. Nahrstedt (University of Illionois at Urbana-Champaign)

  (Abstract requested)

• Light Simulator for Ad-Networks (LSAN)
  L. Cheng, R. Ocampo, A. Galis R. Szabo, C. Simon, P. Kersch (Haifa University)

  (Abstract requested)

• MultiMesh: Multi-Channel Multi-Interface Wireless Mesh Testbed
  C. Chereddi, P. Kaysanur, J. so, N.H. Vaidya (University of Illionois at Urbana-Champaign)

  Wireless mesh networks built using commodity wireless interfaces, such as 802.11, typically make use of as many channels as interfaces. Our research has shown that utilizing more channels than interfaces can significantly increase throughput. We have developed a new architecture for utilizing more channels (potentially all) than interfaces for communication. Our architecture comprises of a link layer module which manages interfaces and channels, masking all complexity from user applications. A higher layer protocol interacts with the link layer module to assign serviced channels and set up neighbor routing tables. Our architecture also has the flexibility to implement other higher layer control protocols. During the demo, we will present a testbed implementation of the proposed architecture.

• Ambient Networks in Practice - Instant Media Services for Users on the Move
  M. Kampmann (Ericsson), M. Vorwerk (Ericsson), M. Kleis (Fraunhofer FOKUS), S. Schuetz (NEC Europe Ltd.), S. Schmid (NEC Europe Ltd.), R. Augero (Unviversity of Cantabria)

  This demonstration exhibits the novel concepts of autonomic network composition and adaptive media delivery, which have been developed as part of the EU Framework 6 Project Ambient Networks. A realistic use case serves as a demonstration scenario to illustrate how these core features of an Ambient Network work together in practice, and how they can contribute to improve the end user's experience in a dynamic and mobile network environment. The scenario is based around a Video-on-Demand (VoD) streaming service. It involves a service provider, network operator and mobile user. The selected scenario highlights the demonstrated Ambient Network features: plug and play network organization, customizable and adaptable transport overlays, and smart media routing are orchestrated.

• Janus - architecture for providing flexible access to sensor network resources
  R. Gold (Uppsala University)

  We present the design and implementation of the Janus architecture (Janus is the two-faced Roman god of doorways) for providing flexible access to sensor network resources from Internet-type networks. Our architecture is designed both to support existing sensor network paradigms, and to be a flexible platform for future development both in terms of new functionality and new combinations of existing functionality. By attempting to maximize flexibility through dynamic negotiation, Janus allows interaction with a wide range of existing types of sensor networks and enables both on-demand querying and event notification. In contrast to existing application-specific access techniques, Janus separates the access technique from the sensor network itself. This attribute allows for application-specific access techniques to be deployed dynamically without requiring the sensor network itself to be upgraded.

• Probabilistic Routing in Intermittently Connected Networks using Lego MindStorm Robots
  C. Baraer. L. Birsa, O. Burman, A. Lindgren, K.S. Phanse (Lulea University of Technology)

  This demo shows the operation of a routing protocol in intermittently connected networks using Lego MindStorms robots. The PRoPHET routing protocol has been implemented on the MindStorms platform and is run on several mobile robots and stationary nodes to demonstrate the operation of the protocol in a visually appealing way. Much effort has been put into making sure that people watching the demo can easily see what is happening in the various parts of the network, both through audible and visual signals on the robots as well as control and logging panels on laptops connected to the network that allow detailed overview and control of everything that is happening