Result Details

A Deep Learning Approach to Multipath Component Detection in Power Delay Profiles

ZELENÝ, O.; ZÁVORKA, R.; PROKEŠ, A.; FRÝZA, T.; WOJTUŃ, J.; KELNER, J.; ZIÓŁKOWSKI, C.; CHANDRA, A. A Deep Learning Approach to Multipath Component Detection in Power Delay Profiles. In Proceeding of the 35th International Conference Radioelektronika (RADIOELEKTRONIKA). Hnanice, Czech republic: IEEE, 2025. p. 319-323. ISBN: 979-8-3315-4447-8.
Type
conference paper
Language
English
Authors
Zelený Ondřej, Ing., UREL (FEEC)
Závorka Radek, Ing., UREL (FEEC)
Prokeš Aleš, prof. Ing., Ph.D., UREL (FEEC)
Frýza Tomáš, doc. Ing., Ph.D., UREL (FEEC)
Wojtuń Jarosław
Kelner Jan M., doc. Ing., Ph.D.
Ziółkowski Cezary, doc. Ing., Dr.
Chandra Aniruddha, Dr., Ph.D.
Abstract

Power Delay Profile (PDP) plays a crucial role in
wireless communications, providing information on multipath
propagation and signal strength variations over time. Accurate
detection of peaks within PDP is essential to identify dominant
signal paths, which are critical for tasks such as channel esti
mation, localization, and interference management. Traditional
approaches to PDP analysis often struggle with noise, low
resolution, and the inherent complexity of wireless environments.
In this paper, we evaluate the application of traditional and
modern deep learning neural networks to reconstruction-based
anomaly detection to detect multipath components within the
PDP. To further refine detection and robustness, a framework is
proposed that combines autoencoders and Density-Based Spatial
Clustering of Applications with Noise (DBSCAN) clustering. To
compare the performance of individual models, a relaxed F1
score strategy is defined. The experimental results show that
the proposed framework with transformer-based autoencoder
shows superior performance both in terms of reconstruction and
anomaly detection.

Keywords

signal propagation, channel measurement, power delay profile, multipath components, peak detection, anomaly detection, machine learning, deep learning

URL
Published
2025
Pages
319–323
Proceedings
Proceeding of the 35th International Conference Radioelektronika (RADIOELEKTRONIKA)
Conference
2025 35th International Conference Radioelektronika (RADIOELEKTRONIKA)
ISBN
979-8-3315-4447-8
Publisher
IEEE
Place
Hnanice, Czech republic
DOI
UT WoS
001509603700030
BibTeX
@inproceedings{BUT197942,
  author="Ondřej {Zelený} and Radek {Závorka} and Aleš {Prokeš} and Tomáš {Frýza} and Jarosław {Wojtuń} and Jan M. {Kelner} and Cezary {Ziółkowski} and Aniruddha {Chandra}",
  title="A Deep Learning Approach to Multipath Component Detection in Power Delay Profiles",
  booktitle="Proceeding of the 35th International Conference Radioelektronika (RADIOELEKTRONIKA)",
  year="2025",
  pages="319--323",
  publisher="IEEE",
  address="Hnanice, Czech republic",
  doi="10.1109/RADIOELEKTRONIKA65656.2025",
  isbn="979-8-3315-4447-8",
  url="https://ieeexplore.ieee.org/document/11008404"
}
Departments
Back to top