On the Application of Deep Learning for Precise Indoor Positioning in 6G

TL;DR

This paper presents LocNet, a neural network model that significantly improves indoor positioning accuracy in 6G environments by leveraging AI/ML techniques on CIR and RSRP data, demonstrating high precision and robustness.

Contribution

Introduction of LocNet, a neural network for indoor positioning that achieves 9 cm accuracy and maintains performance despite measurement unavailability and label errors.

Findings

Achieves 9 cm accuracy at 90th percentile with 18 TRPs

Generalizes well with missing TRP data

Robust to ground truth label errors

Abstract

Accurate localization in indoor environments is a challenge due to the Non Line of Sight (NLoS) nature of the signaling. In this paper, we explore the use of AI/ML techniques for positioning accuracy enhancement in Indoor Factory (InF) scenarios. The proposed neural network, which we term LocNet, is trained on measurements such as Channel Impulse Response (CIR) and Reference Signal Received Power (RSRP) from multiple Transmit Receive Points (TRPs). Simulation results show that when using measurements from 18 TRPs, LocNet achieves a 9 cm positioning accuracy at the 90th percentile. Additionally, we demonstrate that the same model generalizes effectively even when measurements from some TRPs randomly become unavailable. Lastly, we provide insights on the robustness of the trained model to the errors in ground truth labels used for training.