Autoconfiguration - Why do we do this?

In recent years, wireless networks became widely accepted as an alternative to wired networks for connecting end-user devices. Standardization and decreasing costs enabled their success in the mass-market and lately the wide-spread deployment in private households. However, such systems usually only bridge the distance between a base station and the end-user device wirelessly. Behind the base station, there is a more or less complex infrastructure based on traditional wired network technology. For example, the access points of a WLAN are generally interconnected by Ethernet cables.

Over the past years, an approach has received a great deal of attention that consistently adopts wireless network technology: mobile ad hoc networks (MANETs). MANETs are decentralised networks, i.e., every node acts as a host as well as a router and, therefore, contributes both to the network architecture and to the routing (multi-hop communication). There are no special nodes whose failure might tear down the network completely. Thus, it is possible to establish MANETs in situations where traditional wired infrastructure is damaged, or does not even exist, and cannot be re-established in a timely manner. Typical application scenarios are the deployment of relief units in disaster areas or of combat units in battlefields.

Despite intensive research effort, MANETs have not become widely accepted by the mass market, yet. This is comprehensible insofar as the above and frequently mentioned application scenarios are very specific and mainly either military-driven or solely applicable in special cases. For example, see Victor Bahl’s Opening Remarks at Mesh Networking Summit 2004. However, the masses demand versatile networks providing them high bandwidth and access to the Internet.

There is a new class of networks fulfilling these requirements, the so-called wireless mesh networks (WMN). WMNs do not have the aim to establish utterly self-sufficient, isolated networks without any pre-existing infrastructure. They may even be integrated into wired networks and easily extend them at low costs. Thereby, neither mobility nor flexibility is fundamentally lost, since wireless mesh networks finally generalize MANETs and, thus, contain them.

So far, efforts were mainly concentrated on developing appropriate routing protocols for MANETs. However, ease of use is one key factor of success for technological innovation. The desirable extensive automatic configuration of WMNs clearly demands novel approaches since even basic prerequisites for the application of conventional solutions are not met.