Beamformed Channel Matrix Positioning using 5G Testbench CSI data with a Deep-Learning Pipeline

Abstract

Within the telecommunications industry, a positioning system for estimating user
equipment (UE) location using purely information available for the basestation has
an enormous number of potential uses. The link between physical position and the
network channel state enables potential positioning systems to function by under-
standing the network channel state dependency on location, using a model-based,
data-based, or a combined approach.
A key exploitable phenomenon linked to position is that of multi-path propaga-
tion, wherein transmissions can arrive from multiple directions to the UE, with a
unique propagation pattern corresponding to a unique environment. In fifth gen-
eration wireless technology (5G), multi-path components are already exploited
for beamforming with massive multiple-input multiple-output (MIMO) technol-
ogy. Basestations therefore have a preexisting pipeline for obtaining beamformed
channel matrices from channel state information (CSI) transmitted by the UE. A
data-driven approach using multi-path propagation phenomena for positioning is
possible through utilizing the already available beamformed channel matrix in the
basestation.
In this thesis the practical data-driven deep-learning approach for UE position-
ing in 5G using beamformed channel matrices is examined. Real-world data is
utilized to judge the applicability of the approach, with measurements done on a
commercial-grade Ericsson 5G testbench in both non-line-of-sight (nLoS) and line-
of-sight (LoS) scenarios. Using a similar approach to other papers in the field, a su-
pervised deep-learning approach is used for instantaneous position estimation. For
improving positioning accuracy through trajectory estimation, a novel approach of
using particle filtering with network ensemble outputs for kernel density estima-
tion of an observation probability density function is proposed. The results show
that using the outlined methods position is possible to estimate in real-world pedes-
trian tests with a mean accuracy of 2-5 meters, even with nLoS conditions and poor
underlying GNSS training data quality

Popular Abstract

In this thesis a novel combined approach for phone positioning in 5G using deep learning and data processing is examined. The proposed method combination results in a mean accuracy of around 2-5 meters, demonstrating a promising path forward to a workable alternative to current navigation systems.

The concept underlying the proposed method is that physical position and the state of the 5G radio connection are linked. Deep-learning algorithms can then learn to understand this link and predict possible locations of all devices connected to the 5G system. By simulating realistic pedestrian movement in the background, a series of predictions can then be used to better pinpoint real user position in real-time.

In practice, this means that with the methods proposed in this thesis, position can be predicted with existing infrastructure using only 5G connection state. Furthermore, the results were obtained through realistic measurements on Ericsson 5G testbench equipment under non-ideal conditions with the 5G transmitter not directly visible, showing the viability of the system in a real-world scenario.

• author: Ráth, Andre
• supervisor: Fredrik Tufvesson; Bo Bernhardsson; Dino Pjanic; Ove Edfors
• organization: Department of Electrical and Information Technology
• course: EITM02 20221
• year: 2022
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: 5G, Deep Learning, Positioning, Channel State Information, CSI, Deep-learning, neural-network, particle-filter, particle, filter, Channel Matrix, Beamformed Channel Matrix, localization, real-world, data-driven, 5G testbench, basestation
• report number: LU/LTH-EIT 2022-892
• language: English
• additional info: Student Paper
• id: 9100498
• date added to LUP: 2022-09-26 13:47:43
• date last changed: 2022-09-26 13:47:43