Game theory based content placement and pricing in connected edge networks

Over the last decade, an exponential growth in media traffic calls for designing new network architectures for future content distribution networks. Researchers have focused on mitigating the effect of the media traffic by caching them in in-network nodes closer to the end users. Caching content as supported by Information-centric networking (ICN) is regarded as a very encouraging solution to diminish the backhaul traffic load as well as the latency. Different centrality metrics have been used in the past to model different caching strategies in ICN. In this thesis, we propose a complete operational framework from content provider's perspective which includes placement and pricing of digital media contents. At first we propose a novel caching strategy based on game theoretic centrality metric to store a content close (here `close' means that minimum number of hops are required by the user(s) to locate and get the content in a sparsely connected, multi-hop network) to the potentially interested user(s). Based on the content access history of BBC iPlayer in UK, we evaluate our proposed content placement solution in two different real life scenarios. The evaluation is performed to measure different metrics like traffic savings, energy savings, etc. It is also shown that the performance our proposed solution outperforms the state-of-the-art centrality measures. Along with the content placement, we provide a content pricing policy also from the content provider's perspective. Using Nash Bargaining Solution (NBS), we propose a pricing policy so that the both users and platform provider maximize benefits, while the load of the server is minimized.     «

Over the last decade, an exponential growth in media traffic calls for designing new network architectures for future content distribution networks. Researchers have focused on mitigating the effect of the media traffic by caching them in in-network nodes closer to the end users. Caching content as supported by Information-centric networking (ICN) is regarded as a very encouraging solution to diminish the backhaul traffic load as well as the latency. Different centrality metrics have been used...     »