BEAMWAVE: Cross-Layer Beamforming and Scheduling for Superimposed Transmissions in Industrial IoT mmWave Networks

TL;DR

BEAMWAVE introduces a cross-layer design for beamforming and scheduling in mmWave IoT networks, enabling efficient superimposed transmissions for industrial applications with improved performance and scalability.

Contribution

It proposes BEAMWAVE, a novel decomposition approach for joint beamforming and scheduling in LDM systems, addressing NP-hard challenges in industrial IoT networks.

Findings

BEAMWAVE achieves near-optimal performance in simulations.

It outperforms existing schemes in scalability and efficiency.

The approach effectively manages superimposed transmissions in dense IoT environments.

Abstract

The omnipresence of IoT devices in Industry 4.0 is expected to foster higher reliability, safety, and efficiency. However, interconnecting a large number of wireless devices without jeopardizing the system performance proves challenging. To address the requirements of future industries, we investigate the cross-layer design of beamforming and scheduling for layered-division multiplexing (LDM) systems in millimeter-wave bands. Scheduling is crucial as the devices in industrial settings are expected to proliferate rapidly. Also, highly performant beamforming is necessary to ensure scalability. By adopting LDM, multiple transmissions can be non-orthogonally superimposed. Specifically, we consider a superior-importance control multicast message required to be ubiquitous to all devices and inferior-importance private unicast messages targeting a subset of scheduled devices. Due to NP-hardness, we propose BEAMWAVE, which decomposes the problem into beamforming and scheduling. Through simulations, we show that BEAMWAVE attains near-optimality and outperforms other competing schemes.