Cost-Effective Cache Deployment in Mobile Heterogeneous Networks

TL;DR

This paper proposes an optimized hierarchical caching strategy in heterogeneous networks to maximize capacity and ensure service quality within a fixed budget, using stochastic geometry analysis.

Contribution

It introduces a cost-effective cache deployment framework with optimal cache sizing and load balancing in two-tier HetNets based on stochastic geometry.

Findings

Optimal cache sizes are threshold-based with respect to budget.

Hierarchical caching improves network capacity and load balancing.

Simulation confirms effectiveness of the proposed deployment scheme.

Abstract

This paper investigates one of the fundamental issues in cache-enabled heterogeneous networks (HetNets): how many cache instances should be deployed at different base stations, in order to provide guaranteed service in a cost-effective manner. Specifically, we consider two-tier HetNets with hierarchical caching, where the most popular files are cached at small cell base stations (SBSs) while the less popular ones are cached at macro base stations (MBSs). For a given network cache deployment budget, the cache sizes for MBSs and SBSs are optimized to maximize network capacity while satisfying the file transmission rate requirements. As cache sizes of MBSs and SBSs affect the traffic load distribution, inter-tier traffic steering is also employed for load balancing. Based on stochastic geometry analysis, the optimal cache sizes for MBSs and SBSs are obtained, which are threshold-based with respect to cache budget in the networks constrained by SBS backhauls. Simulation results are provided to evaluate the proposed schemes and demonstrate the applications in cost-effective network deployment.