Smart Caching in Wireless Small Cell Networks
Global mobile data traffic has been increasing over the past years, one of its major generators being video traffic, as indicated by a recent study by Cisco Inc. A solution to relieve highly-loaded networks from mobile data traffic is caching at the edge. The most popular files are stored at local caches to serve end users' requests directly, which can reduce bandwidth requirements and delay. A promising caching architecture is given by small cell networks in which small base stations (sBSs) are utilized as local caching entities. These sBSs dispose of limited storage capacities. In their caches, they can store a fraction of the available content to serve users in their vicinity via localized communication.
A crucial question for content caching is, which files to store in order to maximize the number of cache hits. This requires information about file popularity. However, file popularity may vary, for example depending on fluctuating types of users in the vicinity of the sBS. Moreover, information about file popularity may not be available a priori. Hence, the optimal content placement has to be learned by the sBS over time. The goal of this project is to develop algorithms that help the sBS to learn smart caching.