On Edge Caching in the Presence of Malicious Users

TL;DR

This paper examines optimal cache placement strategies in heterogeneous networks considering malicious users aiming to increase backhaul congestion, using game theory to analyze system performance and the impact of adversaries.

Contribution

It introduces an adversary-aware cache placement approach and models the interaction as a Stackelberg game to optimize network performance against malicious users.

Findings

Malicious users can significantly increase backhaul load.

Adversary-aware caching improves network resilience.

System parameters influence the effectiveness of cache strategies.

Abstract

In this paper we investigate the problem of optimal cache placement in the presence of malicious mobile users in heterogeneous networks, where small-cell base stations are equipped with caches in order to reduce the overall backhaul load. In particular the malicious users aim at maximizing the congestion of files at the backhaul, i.e., at maximizing the average backhaul rate. For that adversarial model, we derive the achievable average backhaul rate of the heterogeneous network. Moreover, we study the system performance from a game-theoretic perspective, by naturally considering a zero-sum Stackelberg game between the macro-cell base station and the malicious users. We then thoroughly investigate the system performance in the presence of adversaries and we analyze the influence of the system parameters, such as the network topology and the capabilities of the small-cell base stations, on the overall performance of edge caching at the Stackelberg equilibrium. Our results highlight the impact of the malicious users on the overall caching performance of the network and they furthermore show the importance of an adversary-aware content placement at the small-cell base stations.