Tutorial #1 (Tuesday)

Management of highly dynamic, infrastructureless radio networks

Frank Eyermann, Universität der Bundeswehr, Germany

Abstract

Infrastructureless radio networks are the means for communication in all cases in which a communication infrastructure is not present or can not be used. Such situations can be found after natural catastrophes (e.g., Hurricane Katrina), which destroyed the previously existing infrastructure and thus can not be used by first responders. Other examples are humanitarian missions in less developed areas, where no comprehensive communication infrastructure does exist. Last but not least during crisis operations or hot wars the troops in-theater need to provide their own communication infrastructure as of security issues existing infrastructure can not be used.

From these sketched scenarios the key requirements can be derived:

• Communication must be provided ad hoc. Regularly there is no time to build up an infrastructure.
• Communication equipment must be mobile.
• High dynamic considering movement and number of nodes
• Need to protect communication from eavesdropping

Several techniques are used in combination today, building some kind of communication hierarchy:

• Satellite links provide connection to the home country
• HF point-to-point radio links provide long-haul land-based communication links between sites in the mission area
• Radio circuits or modern radio networks provide communication means between the nodes at one site. Only this lowest level can fulfill the above mentioned requirements.

Radio circuits may provide data connection, but no network capability thus communication is only possible between directly reachable nodes. Nodes of radio networks can forward data of other nodes and thus build a mobile ad hoc network (MANet).

Especially the management of networks and devices of this lowest level of the above depicted communication pyramid is challenging and requires partly completely new approaches. Reasons are the high dynamic of the nodes of the network and the scarce bandwidth available.

Mobile ad hoc networks may be a niche as of today. Their dissemination will, however, rise in near future. Three examples should be named:

• Military identified IP-based mobile ad hoc networks as key enabler for the concept of network-centric warfare (NATO) or network-enabled capability (EU), which should lead to information dominance and thus an advantage on the battle field.
• Software-defined radios can emulate different types of radios thus allowing previously incompatible radio systems to connect.
• These devices provide also sufficient processing power for using modern radio encoding schemes allowing for a higher data rate. Data rates of older systems are as low as several kilobits per second.

The tutorial will first introduce to these networks and make the participants familiar with the technologies, their parameters, and the terms used. Especially militarily operated networks have a set of requirements, which do not exist elsewhere (e.g., red-red separation). Typical scenarios are shown undermining the participants understanding. A short outlook is given to the concept of software-defined radio and its benefits in the depicted scenarios.

In the second part the requirements of a management system are described. As already mentioned above because of the special environment of these networks current management concepts can not be directly mapped. Furthermore the management system must combine network and system management in one. A special focus has to be put on security management. In networks with different levels of classified data different trust relationships exists hampering an integrated management. The networks in focus are often federated networks, consisting of nodes from different organizations, e.g., different non-governmental organizations (NGOs) as the Red Cross, military forces of different nations, or different companies, depending on the concrete scenario.

In a third part the concepts and solutions addressing the before mentioned requirements are shown. This will include an architecture for a management system for highly dynamic infrastructureless radio networks, as well as best practice use cases. An outlook is given to the relatively new topic of a Security Management Infrastructure (SMI), addressing a holistic security management over domain boundaries.

As this research topic is quite new and not wide spread, many problems are not yet satisfactorily solved. The tutorial will close with naming some of the most prominent problems inspiring the participants possibly to contribute to this exciting research topic.

Instructor Bio

Frank Eyermann is a Ph.D. student at the Universität der Bundeswehr Munich, Germany. He will finish his thesis on "Auditing of IP carrying multi-domain Service Level Agreements" in a few months. His other fields of research are network security and management of networks and systems. Before, he was a chief programmer at the Germany Airforce Command and Control Information System Programming Center in Birkenfeld, Germany and a system administrator at the same location. He was Local Organizing Co-Chair at the 10th IFIP/IEEE Symposium on Integrated Management 2007 (IM2007).