Title: Fundamental Limits of Caching: Improved Delivery Rate-Cache Capacity Trade-off

Abstract: We consider a centralized coded caching system consisting of a server delivering N popular files, each of size F bits, to K users through an error-free shared link. Each user is equipped with a local cache of capacity MF bits. The contents can be proactively cached into the user caches by the server over low traffic period; however, without the knowledge of the user demands. During the peak traffic period each user requests a single file from the database. The goal is to utilize the user caches in order to minimize the number of bits delivered by the server over a shared link, known as the delivery rate, to satisfy all the user demands. We propose a novel coded caching algorithm for the cache capacity of M=(N-1)/K, when N and K have a common divisor greater than 1, and satisfy 4<=N<K<=3N/2. It is shown that the proposed scheme achieves a smaller delivery rate than the existing coded caching schemes in the literature for 1/K<=M<=N/K when N<K<3N/2, and 1/K<=M<=2N/K when K=3N/2.