Deep-Learning-Based High-Precision Localization With Massive MIMO

High-precision localization and machine learning (ML) are envisioned to be key technologies in future wireless systems. This paper presents an ML pipeline to solve localization tasks. It consists of multiple parallel processing chains, each trained using a different fingerprint to estimate the posit...

Full description

Saved in:
Bibliographic Details
Main Authors: Guoda Tian, Ilayda Yaman, Michiel Sandra, Xuesong Cai, Liang Liu, Fredrik Tufvesson
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Transactions on Machine Learning in Communications and Networking
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10330061/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850035987291308032
author Guoda Tian
Ilayda Yaman
Michiel Sandra
Xuesong Cai
Liang Liu
Fredrik Tufvesson
author_facet Guoda Tian
Ilayda Yaman
Michiel Sandra
Xuesong Cai
Liang Liu
Fredrik Tufvesson
author_sort Guoda Tian
collection DOAJ
description High-precision localization and machine learning (ML) are envisioned to be key technologies in future wireless systems. This paper presents an ML pipeline to solve localization tasks. It consists of multiple parallel processing chains, each trained using a different fingerprint to estimate the position of the user equipment. In this way, ensemble learning can be utilized to fuse all chains to improve localization performance. Nevertheless, a common problem of ML-based techniques is that network training and fine-tuning can be challenging due to the increase in network sizes when applied to (massive) multiple-input multiple-output (MIMO) systems. To address this issue, we utilize a subarray-based approach. We divide the large antenna array into several subarrays, feeding the fingerprints of the subarrays into the pipeline. In our case, such an approach eases the training process while maintaining or even enhancing the performance. We also use the Nyquist sampling theorem to gain insight on how to appropriately sample and average training data. Finally, an indoor measurement campaign is conducted at 3.7GHz using the Lund University massive MIMO testbed to evaluate the approaches. Localization accuracy at a centimeter level has been reached in this particular measurement campaign.
format Article
id doaj-art-5fa1b09779f04d2eba92888a72e851a3
institution DOAJ
issn 2831-316X
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Transactions on Machine Learning in Communications and Networking
spelling doaj-art-5fa1b09779f04d2eba92888a72e851a32025-08-20T02:57:19ZengIEEEIEEE Transactions on Machine Learning in Communications and Networking2831-316X2024-01-012193310.1109/TMLCN.2023.333471210330061Deep-Learning-Based High-Precision Localization With Massive MIMOGuoda Tian0https://orcid.org/0000-0003-2466-4621Ilayda Yaman1https://orcid.org/0000-0002-2416-2077Michiel Sandra2Xuesong Cai3https://orcid.org/0000-0001-7759-7448Liang Liu4https://orcid.org/0000-0001-9491-8821Fredrik Tufvesson5https://orcid.org/0000-0003-1072-0784Department of Electrical and Information Technology, Lund University, Lund, SwedenDepartment of Electrical and Information Technology, Lund University, Lund, SwedenDepartment of Electrical and Information Technology, Lund University, Lund, SwedenDepartment of Electrical and Information Technology, Lund University, Lund, SwedenDepartment of Electrical and Information Technology, Lund University, Lund, SwedenDepartment of Electrical and Information Technology, Lund University, Lund, SwedenHigh-precision localization and machine learning (ML) are envisioned to be key technologies in future wireless systems. This paper presents an ML pipeline to solve localization tasks. It consists of multiple parallel processing chains, each trained using a different fingerprint to estimate the position of the user equipment. In this way, ensemble learning can be utilized to fuse all chains to improve localization performance. Nevertheless, a common problem of ML-based techniques is that network training and fine-tuning can be challenging due to the increase in network sizes when applied to (massive) multiple-input multiple-output (MIMO) systems. To address this issue, we utilize a subarray-based approach. We divide the large antenna array into several subarrays, feeding the fingerprints of the subarrays into the pipeline. In our case, such an approach eases the training process while maintaining or even enhancing the performance. We also use the Nyquist sampling theorem to gain insight on how to appropriately sample and average training data. Finally, an indoor measurement campaign is conducted at 3.7GHz using the Lund University massive MIMO testbed to evaluate the approaches. Localization accuracy at a centimeter level has been reached in this particular measurement campaign.https://ieeexplore.ieee.org/document/10330061/Channel measurementsdeep learninglocalizationmassive MIMO
spellingShingle Guoda Tian
Ilayda Yaman
Michiel Sandra
Xuesong Cai
Liang Liu
Fredrik Tufvesson
Deep-Learning-Based High-Precision Localization With Massive MIMO
IEEE Transactions on Machine Learning in Communications and Networking
Channel measurements
deep learning
localization
massive MIMO
title Deep-Learning-Based High-Precision Localization With Massive MIMO
title_full Deep-Learning-Based High-Precision Localization With Massive MIMO
title_fullStr Deep-Learning-Based High-Precision Localization With Massive MIMO
title_full_unstemmed Deep-Learning-Based High-Precision Localization With Massive MIMO
title_short Deep-Learning-Based High-Precision Localization With Massive MIMO
title_sort deep learning based high precision localization with massive mimo
topic Channel measurements
deep learning
localization
massive MIMO
url https://ieeexplore.ieee.org/document/10330061/
work_keys_str_mv AT guodatian deeplearningbasedhighprecisionlocalizationwithmassivemimo
AT ilaydayaman deeplearningbasedhighprecisionlocalizationwithmassivemimo
AT michielsandra deeplearningbasedhighprecisionlocalizationwithmassivemimo
AT xuesongcai deeplearningbasedhighprecisionlocalizationwithmassivemimo
AT liangliu deeplearningbasedhighprecisionlocalizationwithmassivemimo
AT fredriktufvesson deeplearningbasedhighprecisionlocalizationwithmassivemimo