Link Scheduling in Multi-Transmit-Receive Wireless Networks

TL;DR

This paper addresses the complex problem of link scheduling in multi-transmit-receive wireless networks, proposing efficient algorithms to optimize network capacity and minimize airtime usage.

Contribution

It formulates the link scheduling problem as a linear program and introduces two novel algorithms, HWF and MDF, to efficiently approximate optimal solutions.

Findings

HWF and MDF outperform existing methods in runtime and optimality

The problem is NP-hard due to the need to find all maximal independent sets

Simulation results demonstrate significant capacity improvements

Abstract

This paper investigates the problem of link scheduling to meet traffic demands with minimum airtime in a multi-transmit-receive (MTR) wireless network. MTR networks are a new class of networks, in which each node can simultaneously transmit to a number of other nodes, or simultaneously receive from a number of other nodes. The MTR capability can be enabled by the use of multiple directional antennas or multiple channels. Potentially, MTR can boost the network capacity significantly. However, link scheduling that makes full use of the MTR capability must be in place before this can happen. We show that optimal link scheduling can be formulated as a linear program (LP). However, the problem is NP-hard because we need to find all the maximal independent sets in a graph first. We propose two computationally efficient algorithms, called Heavy-Weight-First (HWF) and Max-Degree-First (MDF) to solve this problem. Simulation results show that both HWF and MDF can achieve superior performance in terms of runtime and optimality.