Cooperation and Accounting in Multi-Hop Cellular Networks

In the last decade a new paradigm in computer networks gained a lot of popularity, especially in the wireless network research community: mobile ad hoc networks. The main difference between this new and the existing conventional networking paradigm lies in the architecture of the network and the role of the different network components. In the conventional paradigm, a person with a computer connects to a pre-established network infrastructure, which consists of dedicated device to transport and deliver information among computers. In ad hoc networks, the pre-existing infrastructure is missing. Instead, the network is created on demand, with the help of each person and her/his computer. Thus, the computers of individuals take the role of the networking devices. The independence from any pre-installed network infrastructure allows these networks to be ad hoc deployable and to operate at low costs. However, the ad hoc communication paradigm also raises many questions, on if which is how to ensure the participation of the individual persons in the network. If the network participants do not cooperate by providing the network service to others, but only want to use the network selfishly, the network ceases to exist. Because many open questions could not be satisfactory answered yet, mobile ad hoc networks are not widely used outside the research community. With multi-hop cellular networks, this situation might change. Multi-hop cellular networks combine the flexibility of mobile ad hoc networks and the reliability of infrastructure-based networks. Usually, they are comprised of an ad hoc as well as a conventional cellular network part. Therefore, these networks can provide access to persons or information located outside an ad hoc network. Because of the increased reachability and their low deployment costs, they appear to be a viable option for wireless Internet service providers. Still most of the open questions from ad hoc networks remain, however some can be addressed with the help of the provider. In this thesis we argue that cooperation among network participants is a fundamental requirement of ad hoc networks and needs to be guaranteed in order to show the viability of multi-hop communications. Therefore, we propose a novel cooperation and accounting architecture for multi-hop cellular networks, which ensures cooperation among network participants. In particular, we specify a charging and rewarding mechanisms for network participants, which makes cooperation a rewarding alternative to selfishness. In order to retain as much as possible of the