The Right Content for the Right Relay in Self-Organizing Delay Tolerant Networks: A Matching Game Perspective

In this paper, we deal with the store-and-forward paradigm for self-organizing Delay Tolerant Networks (DTN). To overcome the decentralized nature and the infrastructureless constraint of such a network, highly distributed design and efficient incentive mechanisms are needed in order to convince relay nodes to disseminate the content. Here, we exhibit a new way to set the store-and-forward scheme based on the emerging matching game theory. This approach serve to match between on one hand different kinds of files generated by a source node and on the second hand relay nodes that may forward these files. In order to make incentive for cooperation, the source node offers a strategic reward to relay nodes that have accepted to forward a given file. Moreover, each relay and file can be defined by a context, i.e. its characteristics. Based on that, the source would prefer maximize the overall delivery probability at the same time as the relay would try to guarantee the highest possible reward while considering its battery status. Our matching-game-based scheme promises an efficient tradeoff between the overall delivery probability and the energy consumption. For practical considerations, we propose an algorithmic solution to achieve a stable matching between the sets of source files and the set of relay stations. Extensive simulations show that our scheme outperforms the legacy two-hop routing and illustrate the impact of preferences of each set involved in the game, and how such a tool can meet a high delivery rate at a reasonable energy budget.