Current Approaches towards Improved Quality-of-Service Provision in Mobile Ad-hoc Networks

With mobile computers and wireless networking hardware becoming widely and increasingly cheaply available, a large body of effort has over the past years gone into incorporating mobility support into the existing, prevailingly wired Internet. Today’s wireless networks, however, are either restricted to some sort of immobile infrastructure (e.g., UMTS), or to direct, single-hop connections (e.g., IEEE 802.11 DCF), or to both (e.g., DECT). In an infrastructure-oriented network, all communication is switched over a fixed, wired backbone. Stations may freely roam within the network’s extent, yet the network itself cannot be relocated. Building the backbone calls for substantial investments in terms of both money and time. A single-hop wireless network has no routing capability, and two stations cannot exchange data unless they are within each other’s radio range. This type of network seldom grows beyond the scale of a local area network. Mobile Ad-hoc Networks (Manets) are multi-hop wireless networks without an explicit backbone, lifting the restriction and the expenditures of an unmovable infrastructure. In Manets, all stations combine the functionality of clients and routers. The network topology is continuously tracked, and routing paths are constantly updated, as stations freely roam about. Manets can be set up quickly and without effort. They are predestined for spontaneous employment at conventions, military engagements, or during disaster-recovery operations. Research on Manets is still emerging, and while the basic routing problem has received much attention, many challenges are yet to be approached. This includes Quality-of-Service issues: Due to the fundamental difference between wireless multi-hop networks and wired networks (perhaps with a wireless “last hop”), existing approaches for wired networks cannot be conferred without adaptation.