Adaptive attention-based model for 5G radio-based outdoor localization
Yaman, Ilayda LU ; Tian, Guoda ; Pjanić, Dino LU ; Tufvesson, Fredrik LU
; Edfors, Ove
LU
; Zhang, Zhengya
and Liu, Liang
LU
(2025)
The Fifty-Ninth Asilomar Conference on Signals, Systems & Computers
In Conference record / Asilomar Conference on Signals, Systems & Computers
p.192-197
- Abstract
- Radio-based localization in dynamic environments, such as urban and vehicular settings, requires systems that efficiently adapt to varying signal conditions and environmental changes. Factors like multipath interference and obstructions introduce different levels of complexity that affect the accuracy of the localization. Although generalized models offer broad applicability, they often struggle to capture the nuances of specific environments, leading to suboptimal performance in real-world deployments. In contrast, specialized models can be tailored to particular conditions, enabling more precise localization by effectively handling domain-specific variations, which also results in reduced execution time and smaller model size. However,... (More)
- Radio-based localization in dynamic environments, such as urban and vehicular settings, requires systems that efficiently adapt to varying signal conditions and environmental changes. Factors like multipath interference and obstructions introduce different levels of complexity that affect the accuracy of the localization. Although generalized models offer broad applicability, they often struggle to capture the nuances of specific environments, leading to suboptimal performance in real-world deployments. In contrast, specialized models can be tailored to particular conditions, enabling more precise localization by effectively handling domain-specific variations, which also results in reduced execution time and smaller model size. However, deploying multiple specialized models requires an efficient mechanism to select the most appropriate one for a given scenario. In this work, we develop an adaptive localization framework that combines shallow attention-based models with a router/switching mechanism based on a single-layer perceptron. This enables seamless transitions between specialized localization models optimized for different conditions, balancing accuracy and computational complexity. We design three low-complex models tailored for distinct scenarios, and a router that dynamically selects the most suitable model based on real-time input characteristics. The proposed framework is validated using real-world vehicle localization data collected from a massive MIMO base station and compared to more general models. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/de9d05a5-b383-464b-a7b9-900285e13f84
- author
- Yaman, Ilayda
LU
; Tian, Guoda
; Pjanić, Dino
LU
; Tufvesson, Fredrik
LU
; Edfors, Ove
LU
; Zhang, Zhengya
and Liu, Liang
LU
- organization
- publishing date
- 2025
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- Conference Record of The Fifty-Ninth Asilomar Conference on Signals, Systems & Computers
- series title
- Conference record / Asilomar Conference on Signals, Systems & Computers
- editor
- Matthews, Michael B.
- pages
- 192 - 197
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- The Fifty-Ninth Asilomar Conference on Signals, Systems & Computers
- conference location
- Pacific Grove, United States
- conference dates
- 2025-10-26 - 2025-10-29
- external identifiers
-
- scopus:105035826329
- ISSN
- 1058-6393
- ISBN
- 979-8-3315-8745-1
- 979-8-3315-8746-8
- DOI
- 10.1109/IEEECONF67917.2025.11443733
- language
- English
- LU publication?
- yes
- id
- de9d05a5-b383-464b-a7b9-900285e13f84
- date added to LUP
- 2026-04-08 10:26:51
- date last changed
- 2026-05-25 04:00:58
@inproceedings{de9d05a5-b383-464b-a7b9-900285e13f84,
abstract = {{Radio-based localization in dynamic environments, such as urban and vehicular settings, requires systems that efficiently adapt to varying signal conditions and environmental changes. Factors like multipath interference and obstructions introduce different levels of complexity that affect the accuracy of the localization. Although generalized models offer broad applicability, they often struggle to capture the nuances of specific environments, leading to suboptimal performance in real-world deployments. In contrast, specialized models can be tailored to particular conditions, enabling more precise localization by effectively handling domain-specific variations, which also results in reduced execution time and smaller model size. However, deploying multiple specialized models requires an efficient mechanism to select the most appropriate one for a given scenario. In this work, we develop an adaptive localization framework that combines shallow attention-based models with a router/switching mechanism based on a single-layer perceptron. This enables seamless transitions between specialized localization models optimized for different conditions, balancing accuracy and computational complexity. We design three low-complex models tailored for distinct scenarios, and a router that dynamically selects the most suitable model based on real-time input characteristics. The proposed framework is validated using real-world vehicle localization data collected from a massive MIMO base station and compared to more general models.}},
author = {{Yaman, Ilayda and Tian, Guoda and Pjanić, Dino and Tufvesson, Fredrik and Edfors, Ove and Zhang, Zhengya and Liu, Liang}},
booktitle = {{Conference Record of The Fifty-Ninth Asilomar Conference on Signals, Systems & Computers}},
editor = {{Matthews, Michael B.}},
isbn = {{979-8-3315-8745-1}},
issn = {{1058-6393}},
language = {{eng}},
pages = {{192--197}},
publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
series = {{Conference record / Asilomar Conference on Signals, Systems & Computers}},
title = {{Adaptive attention-based model for 5G radio-based outdoor localization}},
url = {{http://dx.doi.org/10.1109/IEEECONF67917.2025.11443733}},
doi = {{10.1109/IEEECONF67917.2025.11443733}},
year = {{2025}},
}