2018/03/11 by Jaber Kakar, Kakar, Jaber, Alameer, Alaa +6
Computer Science · #Caching and Content Delivery #Cooperative Communication and Network Coding #FOS: Computer and information sciences #Information Theory (cs.IT) #Wireless Networks and Protocols
paper · pdf · doi:10.48550/arxiv.1803.04058
openalex publication_date 2018/03/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
An emerging trend of next generation communication systems is to provide\nnetwork edges with additional capabilities such as storage resources in the\nform of caches to reduce file delivery latency. To investigate this aspect, we\nstudy the fundamental limits of a cache-aided broadcast-relay wireless network\nconsisting of one central base station, M cache-equipped transceivers and K\nreceivers from a latency-centric perspective. We use the normalized delivery\ntime (NDT) to capture the per-bit latency for the worst-case file request\npattern, normalized with respect to a reference interference-free system with\nunlimited transceiver cache capabilities. The objective is to design the\nschemes for cache placement and file delivery in order to minimize the NDT. To\nthis end, we establish a novel converse and two types of achievability schemes\napplicable to both time-variant and invariant channels. The first scheme is a\ngeneral one-shot scheme for any M and K that synergistically exploits both\nmulticasting (coded) caching and distributed zero-forcing opportunities. We\nshow that the proposed one-shot scheme (i) attains gains attributed to both\nindividual and collective transceiver caches (ii) is NDT-optimal for various\nparameter settings, particularly at higher cache sizes. The second scheme, on\nthe other hand, designs beamformers to facilitate both subspace interference\nalignment and zero-forcing at lower cache sizes. Exploiting both schemes, we\nare able to characterize for various special cases of M and K which satisfy\nK+M\≤ 4 the optimal tradeoff between cache storage and latency. The\ntradeoff illustrates that the NDT is the preferred choice to capture the\nlatency of a system rather than the commonly used sum degrees-of-freedom (DoF).\nIn fact, our optimal tradeoff refutes the popular belief that increasing cache\nsizes translates to increasing the achievable sum DoF.\n