A Mean Field Game Approach to Scheduling in Cellular Systems

TL;DR

This paper models auction-based scheduling in cellular networks using mean field game theory, demonstrating that auctions can effectively schedule apps with long queues and converge to equilibrium with simple computation.

Contribution

It introduces a mean field game framework for auction-based scheduling in cellular systems, showing existence of equilibrium and convergence properties.

Findings

Existence of a mean field equilibrium for auction-based scheduling.

Auctions can replicate queue-length-based scheduling outcomes.

Finite agent systems converge to the mean field equilibrium asymptotically.

Abstract

We study auction-theoretic scheduling in cellular networks using the idea of mean field equilibrium (MFE). Here, agents model their opponents through a distribution over their action spaces and play the best response. The system is at an MFE if this action is itself a sample drawn from the assumed distribution. In our setting, the agents are smart phone apps that generate service requests, experience waiting costs, and bid for service from base stations. We show that if we conduct a second-price auction at each base station, there exists an MFE that would schedule the app with the longest queue at each time. The result suggests that auctions can attain the same desirable results as queue-length-based scheduling. We present results on the asymptotic convergence of a system with a finite number of agents to the mean field case, and conclude with simulation results illustrating the simplicity of computation of the MFE.