Optimal Geographical Caching in Heterogeneous Cellular Networks

TL;DR

This paper studies optimal content placement in heterogeneous cellular networks with different base station types and cache sizes, proposing convex optimization solutions for PPP deployments and a distributed algorithm for general cases.

Contribution

It introduces a convex formulation for joint caching optimization in PPP-deployed networks and a distributed local optimization algorithm for more general scenarios.

Findings

Optimal placement in PPP networks is convex and solvable.

Storing the most popular content in macro BSs is nearly optimal for small BSs.

The distributed algorithm achieves the same hit probability as the joint optimal solution.

Abstract

We investigate optimal geographical caching in heterogeneous cellular networks, where different types of base stations (BSs) have different cache capacities. The content library contains files with different popularities. The performance metric is the total hit probability. The problem of optimally placing content in all BSs jointly is not convex in general. However, we show that when BSs are deployed according to homogeneous Poisson point processes (PPP), independently for each type, we can formulate the problem as a convex problem. We give the optimal solution to the joint problem for PPP deployment. For the general case, we provide a distributed local optimization algorithm (LOA) that finds the optimal placement policies for different types of BSs. We find the optimal placement policy of the small BSs (SBSs) depending on the placement policy of the macro BSs (MBSs). We show that storing the most popular content in the MBSs is almost optimal if the SBSs are using an optimal placement policy. Also, for the SBSs no such heuristic can be used; the optimal placement is significantly better than storing the most popular content. Finally, we numerically verify that LOA gives the same hit probability as the joint optimal solution for the PPP model.