Context-Aware Wireless Small Cell Networks: How to Exploit User Information for Resource Allocation

TL;DR

This paper introduces a context-aware resource allocation method for small cell networks that leverages user behavior to improve performance, demonstrating significant gains over traditional approaches through game-theoretic modeling.

Contribution

It proposes a novel game-theoretic resource allocation framework that exploits user context information, ensuring improved utility for small cell networks with proven equilibrium properties.

Findings

Significant performance improvements over conventional methods.

Existence and uniqueness of Nash equilibrium independent of user count.

Effective exploitation of user context enhances resource allocation efficiency.

Abstract

In this paper, a novel context-aware approach for resource allocation in two-tier wireless small cell networks~(SCNs) is proposed. In particular, the SCN's users are divided into two types: frequent users, who are regular users of certain small cells, and occasional users, who are one-time or infrequent users of a particular small cell. Given such \emph{context} information, each small cell base station (SCBS) aims to maximize the overall performance provided to its frequent users, while ensuring that occasional users are also well serviced. We formulate the problem as a noncooperative game in which the SCBSs are the players. The strategy of each SCBS is to choose a proper power allocation so as to optimize a utility function that captures the tradeoff between the users' quality-of-service gains and the costs in terms of resource expenditures. We provide a sufficient condition for the existence and uniqueness of a pure strategy Nash equilibrium for the game, and we show that this condition is independent of the number of users in the network. Simulation results show that the proposed context-aware resource allocation game yields significant performance gains, in terms of the average utility per SCBS, compared to conventional techniques such as proportional fair allocation and sum-rate maximization.