Multi-Band mm-Wave Measurement Platform Towards Environment-Aware Beam Management
Aleksandar Ichkov, Aron Schott, Niklas Beckmann, Ljiljana Simić
TL;DR
The paper tackles the problem of agile mm-wave beam management under site-specific channel dynamics by leveraging environment awareness from non-RF sensors to supplement RF-based steering. It presents a software-defined radio based, multi-band measurement platform that integrates LiDAR, camera, and GPS inputs with RF front-ends to enable environment-aware beam management and dataset generation. Key contributions include a packetized TX/RX mm-wave link operating in the 28 GHz and 60 GHz bands with up to 400 MHz bandwidth, support for simultaneous multi-band transmissions, and a combination of electronic and mechanical beam steering. The platform provides synchronized, multi-modal data collection and a plan to release GNU Radio code to enable open, reproducible mm-wave datasets and robust evaluation of environment-aware beam management.
Abstract
Agile beam management is key for providing seamless millimeter wave (mm-wave) connectivity given the site-specific spatio-temporal variations of the mm-wave channel. Leveraging non radio frequency (RF) sensor inputs for environment awareness, e.g. via machine learning (ML) techniques, can greatly enhance RF-based beam steering. To overcome the lack of diverse publicly available multi-modal mm-wave datasets for the design and evaluation of such novel beam steering approaches, we demonstrate our software-defined radio multi-band mm-wave measurement platform which integrates multi-modal sensors towards environment-aware beam management.