Optimal coded caching in heterogeneous networks with uncoded prefetching

In the context of caching in heterogeneous networks, the work explores the setting where a multi-antenna transmitter (N0 antennas), broadcasts to K receiving users, each assisted by one of   K helper nodes serving as limited-sized caches. Our aim is to identify the limits of coded caching when there are fewer caches than users ( < K), the interplay between having fewer caches but more transmit antennas, and the impact of non-uniformity where some caches serve more users than others. For a broad range of parameters, under the assumption of uncoded cache placement, the work derives the exact optimal worst-case delivery time (or equivalently, the optimal sum degrees of freedom (DoF)), as a function of the cache sizes and the user-to-cache association profile. This is achieved by presenting an information-theoretic outer bound based on index coding that adapts to user-to-cache association non-uniformities, and an optimal caching-and-delivery scheme. The result reveals the effect of these non-uniformities, and also reveals a powerful effect of introducing a modest number of antennas and a modest number of helper nodes; when  < K=N0, adding a single degree of cache-redundancy yields a caching-gain increase equal to N0, and similarly, adding antennas has a multiplicative DoF impact where for example introducing a second transmit antenna can double the DoF.