Mechanisms for Secure Charging in Self-Organised Networks

Abstract

Self-organised and ad-hoc networks are an area with an existing large research community. These networks are much useful in scenarios requiring a rapidly deployed, low cost and highly adaptable network. Recently, infrastructure networks, which are managed in a much centralised form, are starting to introduce concepts of self-organised networks in its architecture. In opposition to centralised systems, self-organisation creates the necessity for all nodes to behave according to the best interest of the network. The fact that in many ad-hoc networks nodes have scarce resources poses some threats to this requirement. As resources decreases, such as battery or wireless bandwidth, nodes can start acting selfishly. This behaviour is known to bring damage to self-organised networks and threatens the entire network. Several proposals were made in order to promote the correct usage of the network. Some proposals are based on local information and direct credit exchange while others envision the existence of a central bank. The later solutions are further elaborated in this thesis, as they make possible integration of ad-hoc network with operator driven infrastructures. This work presents the current state-of-the-art on the area providing a detailed insight on the methods adopted by each solution presented. Two novel solutions are proposed providing charging support for integrated ad-hoc networks. Both solutions provide means of integration with standard management methods found in operator networks. Also, node’s motivation is increased through the reward of nodes forwarding data packets. The entire process is cryptographically secure, making use of standard methods such as Elliptic Curve DSA and strong digest functions. The solutions proposed are described and analysed analytically, comparing the results with other state-of-the-art proposals. Extensive simulation work is also presented which furthers evaluates the solutions in complex scenarios. Results are obtained from these scenarios and several metrics are evaluated taking in consideration mobility, network load, routing protocol and transport protocol. The architecture and results obtained with a real implementation are finally presented and analysed.

Type