Game Theoretic Research for Cache Allocation among Base Stations

• 1、School of Electronic Engineering, Beijing University of Posts and Telecommunications, Beijing 100876

Abstract：The intelligent base station network in mobile edge computing environment generally consists of several base stations (MEC-BS) equipped with mobile edge computing server. How to make full use of the limited storage capacity of MEC-BS and improve the cache-hit ratio is the core technical problem in this field. In this paper, a game theory-based cache allocation technique for maximizing utility among MEC-BSs is proposed. This technology considers a heterogeneous network consisting of network service providers (NSP), popular content providers (PCPs), and MEC-BSs. A caching model is actually established which is based on the users' demonds for popular content files and the storage constraints of MEC-BS. Different profit models are built in this heterogeneous network and the research problem is defined as a Starkberg game. Finally, an asynchronous algorithm is designed to simulate the non-cooperative game, and the fast convergence of the proposed algorithm and the theoretical correctness of the concavity of the leader's utility function are analyzed by simulation.

Keywords： Mobile Edge Computing Cache Allocation Game Theory Asynchronous Algorithm