SFP: Statistical Filtering Policy for Caching in Content-Centric Networking

Content-Centric Networking (CCN) is a new paradigm where caching techniques play an important role. This is motivated by the addition of memory to core routers allowing to store the content traversing them. In this environment, several techniques have been proposed to increase the performance of the cache, and most of them are based on well-known ideas that are being used nowadays. In this paper, we propose a new cache decision policy, which is based on statistical filtering by using Bloom filters. Our proposal efficiently improves the hit probability of a CCN router taking into account the popularity and size of the catalogue, and this improvement allow clients benefitting of better download times and lower delays, and also reducing the server load. Furthermore, this method of filtering content takes advantage of techniques that have not been yet explored in this research field. We propose a mathematical model to characterize the decision policy, validating it by means of simulation