Wireless Network Stability in the SINR Model

TL;DR

This paper investigates wireless network stability under stochastic packet arrivals using the SINR interference model, proposing distributed algorithms that achieve near-optimal efficiency across various network configurations.

Contribution

It introduces new distributed algorithms that guarantee stability with provable efficiency bounds in the SINR model for different power assignment strategies.

Findings

Achieves $rac{1}{ ext{log}^2 n}$-efficiency for all networks with monotone, sub-linear power.

Provides a $rac{1}{ ext{log} n ( ext{log} n + ext{log} ext{log} riangle)}$-efficiency algorithm for power control.

Demonstrates stability under stochastic packet arrivals with efficient distributed methods.

Abstract

We study the stability of wireless networks under stochastic arrival processes of packets, and design efficient, distributed algorithms that achieve stability in the SINR (Signal to Interference and Noise Ratio) interference model. Specifically, we make the following contributions. We give a distributed algorithm that achieves $\Omega(\frac{1}{\log^2 n})$-efficiency on all networks (where $n$ is the number of links in the network), for all length monotone, sub-linear power assignments. For the power control version of the problem, we give a distributed algorithm with $\Omega(\frac{1}{\log n(\log n + \log \log \Delta)})$-efficiency (where $\Delta$ is the length diversity of the link set).