Radio resource sharing and edge caching with latency constraint for local 5G operator : geometric programming meets Stackelberg game

Abstract

<div lang="en" class="abs"><p class="abs_header">Abstract</p><p>We develop a novel game-theoretic framework with geometric programming to model and analyze cache-enabled base stations (BSs) with infrastructure sharing for local 5G operator (OP) networks. In such a network, the local 5G OP provides wireless network in indoor area and rents out the infrastructure which are RAN and cache storage to multiple mobile network operators (MNOs) while guarantee the quality-of-experience (QoE) at the users (UEs) of MNOs. We formulate a Stackelberg game model where the local 5G OP is the leader and the MNOs are the followers. The local 5G OP aims to maximize its profit by optimizing its infrastructure rental fee, and the MNOs aim to minimize their renting cost of infrastructure by minimizing the “cache intensity” subject to latency constraint at each UE. The optimization problems of the local 5G OP and the MNOs are transformed into geometric programming. Accordingly, the Stackelberg equilibrium is obtained through the succesive geometric programming method. Since the MNOs share their rented infrastructure, for cost sharing, we apply the concept of Shapley value to divide the cost among the MNOs. Finally, we present an extensive performance evaluation that reveals interesting insights into designing resource sharing with edge caching in local 5G OP networks.</p></div>