Low Complexity Position Estimator for DL-AoD and UL-AoA for B5G and 6G Networks [Runner up: Best Paper Award]

Abstract

6G networks aim to support the ambitious use cases outlined in the ITU IMT 2030 vision document. These use cases demand inherent support for precise localization and accurate sensing capabilities, necessitating advancements in angle, time, and phase measurement capabilities in the present networks and devices. The positioning and sensing requirements of 6G networks can be met by systems that support extreme massive MIMO, wide bandwidth, improved link budgets, and enhanced signal processing for positioning methods. This work proposes an angle-based position estimation method in a massive MIMOOFDM system capable of achieving centimeter-level positioning accuracy within the hardware and computational constraints. The method provides improved performance for both downlink angle of departure (DL-AoD) and uplink angle of arrival (UL-AoA) techniques. Additionally, the proposed method satisfies the key performance indicators (KPIs) of 20 cm and 3 m positioning accuracy for the 90 percentile of user equipment (UEs) for outdoor and indoor factory scenarios respectively as required by releases 16 and 17 specifications of 3GPP. Furthermore, it outperforms traditional iterative algorithms (such as gradient descent and Newton-Raphson) by more than 2× in terms of 90 percentile positioning accuracy, while incurring three to four orders lower computational complexity and 50 percent lower memory complexity.