Map-Assisted Millimeter Wave and Terahertz Position Location and Sensing

Abstract

The vast bandwidth available at millimeter wave (mmWave) and terahertz (THz) frequencies will allow future 6G wireless networks to support ubiquitous and extremely accurate localization and environmental sensing. Prior geometric localization algorithms typically assume single bounce reflections. This paper describes map-assisted positioning with angle and time (MAP-AT), a novel map-based localization algorithm that takes into account multi-bounce reflections, utilizing the angle of arrival and time of flight of multipath signal components to determine the position of a user. The accuracy of MAP-AT is tested against indoor and factory measurement data at mmWave (28 GHz, 60 GHz) and sub-THz (140 GHz) frequencies. Using a single base station as reference, sub-meter accuracy was achieved at mmWave frequencies, and centimeter-level accuracy was achieved at sub-THz frequencies. Accuracy was improved when more base stations were used. Additionally, the performance of sub-Thz signals for sensing objects behind walls is studied by detecting hidden objects behind plywood and drywall in a laboratory environment, with centimeter-level sensing accuracy and identification of hidden objects successfully achieved. This work shows that the high penetration loss of walls and obstructions at sub-THz frequencies poses a challenge to accurate sensing at sub-THz frequencies. Future work is required to sense objects hidden tens of meters behind walls.