Delay Analysis and Optimization in Cache-enabled Multi-Cell Cooperative Networks

TL;DR

This paper models and optimizes cache-enabled multi-cell cooperative networks to minimize delay, introducing a novel delay expression, formulating an NP-complete problem, and proposing efficient greedy caching algorithms with proven performance.

Contribution

It provides a new stochastic delay model, formulates the caching optimization as a submodular maximization problem, and develops improved greedy algorithms for delay reduction.

Findings

Analytical delay expressions match simulation results.

Proposed greedy algorithms outperform conventional methods.

Significant delay reduction demonstrated in simulations.

Abstract

Caching at the base stations (BSs) has been widely adopted to reduce the delivery delay and alleviate the backhaul traffic between BSs and the core network. In this paper, we consider a collaborative content caching scheme among BSs in cache-enabled multi-cell cooperative networks, where the requested contents can be obtained from the associated BS, the other collaborative BSs or the core network. Novelly, we model the stochastic request traffic and derive a closed form expression for the average delay per request based on multi-class processor sharing queuing theory. We then formulate a cooperative caching optimization problem of minimizing the average delay under the finite cache size constraint at BSs and show it to be at least NP-complete. Furthermore, we prove it equivalent to the maximization of a monotone submodular function subject to matroid constraints, allowing us to adopt the common greedy algorithm with 1/2 performance guarantee. A heuristic greedy caching strategy is also developed, achieving a better performance than the conventional greedy solution. Simulation results verify the accuracy of the analytical results and demonstrate the performance gains obtained by our proposed caching scheme.