Caching in Wireless Small Cell Networks: A Storage-Bandwidth Tradeoff

TL;DR

This paper analyzes the tradeoffs between storage capacity and bandwidth in wireless small cell networks, providing a mathematical model for cache-miss probability and demonstrating the benefits of popularity-based caching strategies.

Contribution

It offers a closed-form expression for cache-miss probability considering various network parameters and compares the effectiveness of popularity-based versus uniform caching strategies.

Findings

Popularity-based caching reduces cache-miss probability compared to uniform caching.

Optimal caching strategies improve network performance in terms of outage probability.

Analytical expressions enable better design of cache deployment in small cell networks.

Abstract

Caching contents at the network edge is an efficient mean for offloading traffic, reducing latency and improving users' quality-of-experience. In this letter, we focus on aspects of storage-bandwidth tradeoffs in which small cell base stations are distributed according to a homogeneous Poisson point process and cache contents according to a given content popularity distribution, subject to storage constraints. We provide a closed-form expression of the cache-miss probability, defined as the probability of not satisfying users' requests over a given coverage area, as a function of signal-to-interference ratio, cache size, base stations density and content popularity. In particular, it is shown that for a given minimum cache size, the popularity based caching strategy achieves lower outage probability for a given base station density compared to uniform caching. Furthermore, we show that popularity based caching attains better performance in terms of cache-miss probability for the same amount of spectrum.