September 22 – 25 , 2013, Dagstuhl Seminar 13392
Inter-Vehicular Communication - Quo Vadis
Onur Altintas (TOYOTA InfoTechnology Center – Tokyo, JP)
Falko Dressler (Universität Innsbruck, AT)
Hannes Hartenstein (KIT – Karlsruher Institut für Technologie, DE)
Ozan K. Tonguz (Carnegie Mellon University – Pittsburgh, US)
1 / 2 >
For support, please contact
- Wenn ein Auto mit dem anderen spricht
Article about seminar 13392, published in the Saarbrücker Zeitung on September 24, 2013 (in German).
- Mehr Sicherheit im Straßenverkehr durch kommunizierende Autos Press Release (in German)
The management and control of network connections among vehicles and between vehicles and an existing network infrastructure is currently one of the most challenging research fields in the networking domain. Using the terms Vehicular Ad-hoc Networks (VANETs), Inter-Vehicle Communication (IVC), Car-2-X (C2X), or Vehicle-2-X (V2X), many applications - as interesting as challenging -- have been envisioned and (at least) partially realized. In this context, a very active research fields has developed. There is a long list of desirable applications that can be grouped into four categories:
- eSafety applications that try to make driving safer, e.g., road hazard warning;
- traffic efficiency applications aiming at more efficient and thus greener traffic, e.g., detection of traffic jams;
- manufacturer oriented applications, e.g., automatic software updates; and
- comfort and entertainment applications, e.g., automatic map updates or video streaming.
While there are some similarities with fields like mobile ad-hoc networks or wireless sensor networks, the specific characteristics of vehicular networks require different communication paradigms, different approaches to security and privacy, or different wireless communication systems. For example, the nodes usually do not have severe power and form factor constraints, and they might be always on. On the other hand, due to high relative speeds, wireless connections may not be stable for a longer time period and the network density is expected to vary from sparse to very dense networks. Another challenging issue is the efficient use of available infrastructure, such as road side units or even cellular networks. Furthermore, IVC has strong links to other research domains, e.g., geo-informatics as it requires very precise localization and precise maps or highly scalable simulations that are a requirement for analyzing traffic systems with hundreds or thousands of vehicles.
In the past, many specific solutions for IVC have been identified and now, industry and other stake-holders are already calling for standardization. Still, we believe that many important research questions have only been partially answered and the approaches discussed in the standardization bodies are based only on a minimum consensus of simplest solutions. Security and privacy, scalability, use of advanced communication patterns like aggregation, transmit power control, and optimal medium access are just a few of such issues.
In 2010, a first Dagstuhl Seminar (10402) was organized on the topic of inter-vehicular communication [1,2]. The motivation was to bring together experts in this field to investigate the state of the art and to highlight where sufficient solutions already existed. The main outcome of this very inspiring seminar was that there are indeed areas within this research where scientific findings are being consolidated and adapted by industry. This was the consensus of quite intriguing discussions among participants from both industry and academia. Yet, even more aspects have been identified where substantial research is still needed. These challenges have been summarized in the Dagstuhl report and in the following IEEE Communications Magazine article :
It was the goal of this new seminar to again bring together leading researchers both from academia and industry to discuss if and where the previously identified challenges have been adequately addressed, and to highlight where sufficient solutions exist today, where better alternatives need to be found, and also to give directions where to look for such alternatives. Furthermore, the goal of this workshop was to go on step beyond and identify where IVC can contribute to the basic foundations of computer science or where previously unconsidered foundations can contribute to IVC.
The 2010 Dagstuhl seminar promoted a "top-down" approach to inter-vehicle communications instead of the classical "bottom-up" approach. With the top-down approach, the effects of applications are first analyzed under the assumption that the communication system will be able to support the application. Thus, an "upper bound" can be presented on the benefits of IVC. In our discussions, we summarized all the scientific work that followed this approach after the previous Dagstuhl seminar and contrasted it with new insights based on field operational tests, safety application design and massively distributed operations.
In particular, we shifted the focus from basic networking principles to applicability in real world scenarios. In the last few years, first field operational tests have been conducted in the US (the Michigan field trial) as well as in Europe (SIM-TD in Germany, DRIVE C2X in Europe). Lessons learned from those tests applied to currently used models and concepts will bring new insights into the forthcoming research challenges. Among others, questions to be studied include the following still unanswered research challenges:
- Data analysis of current field operational tests: are they validating or invalidating current models?
- Safety applications: show stopper or driving force? What are the limitations in terms of latency and reliability of available communication principles for enabling critical safety support;
- From highly distributed to massively distributed operation: can vehicular networking based on DSRC/WAVE also support all the pedestrians and bicyclists?
We organized the 2013 seminar again as a discussion forum. Three invited keynote presentations were organized to stimulate discussions among the participants. In order to steer the discussions, we prepared four working groups that helps focusing on selected open research challenges. In addition, we also supported ad-hoc presentations on topics of the working groups. The following working groups have been formed and led to very interesting observations:
- Foundations -- In this group, it was discussed, which fundamental insights gained in the vehicular networking research domain can be transfered to other domains of computer science. The other way around has been discussed as well, i.e., which areas of computer science might help fostering work in the vehicular networking and which may help overcoming open challenges.
- Field Operational Tests (FOTs) -- This group focused on the results that already have been derived from the ongoing work in various test sites in the U.S. and in Europe. The main questions in the discussion were whether the current experiments are already sufficient to gain insights into larger scale behavior or if additional tests are needed.
- IVC Applications -- In this group, the applications' perspective to IVC was discussed. In the last years, many of the developments have been done looking at lower layer networking problems. This resulted in a number of networking solutions that nicely support specific applications but cannot be integrated to a generalized networking architecture.
- Heterogeneous Networks -- Possibly one of the most important and timely working groups focused on the integration of different networking technologies. This is strongly needed to develop integrated IVC solutions and also to overcome early deployment problems like the initially low penetration ratio.
Eventually, all these questions lead to the big question whether vehicular networking can now be shown to improve efficiency and safety on our streets. We are now in an era that completely changes the game in car manufacturing and road traffic management. Computer science is becoming the key element in the design of these systems. It is of utmost importance to bring in expertise from classical computer science (computer networking, simulation and modeling, operating system design) as well as from electrical engineering (digital signal processing, communication networks) as well as experts from the automotive industry and from the intelligent transportation community.
- F. Dressler, F. Kargl, J. Ott, O. K. Tonguz, and L. Wischhof, "Executive Summary - Inter-Vehicular Communication", in Dagstuhl Seminar 10402 - Inter-Vehicular Communication. Schloss Dagstuhl, Wadern, Germany: Schloss Dagstuhl, October 2010. Available: http://drops.dagstuhl.de/opus/volltexte/2011/2929/
- Falko Dressler, Frank Kargl, Jörg Ott, Ozan K. Tonguz and Lars Wischhof, "Research Challenges in Inter-Vehicular Communication - Lessons of the 2010 Dagstuhl Seminar," IEEE Communications Magazine, vol. 49 (5), pp. 158-164, May 2011.
Creative Commons BY 3.0 Unported license
Onur Altintas and Falko Dressler and Hannes Hartenstein and Ozan K. Tonguz
Related Dagstuhl Seminar
- 10402: "Inter-Vehicular Communication" (2010)
- Mobile Computing
- Intelligent transportation systems
- Vehicle-to-vehicle communication
- Vehicle-to-infrastructure communication
- Vehicular ad hoc networks
- Traffic information system
- Traffic flow optimization
- Energy and emission efficiency