Distributed Learning for Proportional-Fair Resource Allocation in Coexisting WiFi Networks

TL;DR

This paper proposes a distributed algorithm enabling WiFi nodes to independently tune their contention windows for proportional fairness, improving network utility and throughput in coexisting WiFi networks with diverse modulation schemes.

Contribution

It introduces a simple, distributed learning-based algorithm for proportional-fair resource allocation, building on recent theoretical work on asynchronous distributed optimization.

Findings

The proposed algorithm achieves better throughput than standard WiFi back-off.

It effectively mitigates the performance anomaly caused by diverse modulation schemes.

The method enables independent tuning of contention windows by WiFi nodes.

Abstract

In this paper, we revisit the widely known performance anomaly that results in severe network utility degradation in WiFi networks when nodes use diverse modulation and coding schemes. The proportional-fair allocation was shown to mitigate this anomaly and provide a good throughput to the stations. It can be achieved through the selection of contention window values based on the explicit solution of an optimization problem or, as proposed recently, by following a learning-based approach that uses a centralized gradient descent algorithm. In this paper, we leverage our recent theoretical work on asynchronous distributed optimization and propose a simple algorithm that allows WiFi nodes to independently tune their contention window to achieve proportional fairness. We compare the throughputs and air-time allocation that this algorithm achieves to those of the standard WiFi binary exponential back-off and show the improvements.