Estimating Multipath Component Delays with Transformer Models
Ott J, Stahlke M, Feigl T, Mutschler C (2024)
Publication Type: Journal article
Publication year: 2024
Journal
Pages Range: 1-10
DOI: 10.1109/JISPIN.2024.3422908
Abstract
Multipath in radio propagation provides essential environmental information that is exploited for positioning or channel-simultaneous localization and mapping. This enables accurate and robust localization that requires less infrastructure than traditional methods. A key factor is the reliable and accurate extraction of multipath components (MPC). However, limited bandwidth and signal fading make it difficult to detect and determine the parameters of the individual signal components. In this article, we propose multipath delay estimation based on a Transformer (TF) neural network. In contrast to the state-of-the-art, we implicitly estimate the number of MPCs and achieve subsample accuracy without using computationally intensive super-resolution techniques. Our approach outperforms known methods in detection performance and accuracy at different bandwidths. Our ablation study shows exceptional results on simulated and real datasets and generalizes to unknown radio environments.
Authors with CRIS profile
Involved external institutions
Germany (DE)How to cite
APA:
Ott, J., Stahlke, M., Feigl, T., & Mutschler, C. (2024). Estimating Multipath Component Delays with Transformer Models. IEEE Journal of Indoor and Seamless Positioning and Navigation, 1-10. https://doi.org/10.1109/JISPIN.2024.3422908
MLA:
Ott, Jonathan, et al. "Estimating Multipath Component Delays with Transformer Models." IEEE Journal of Indoor and Seamless Positioning and Navigation (2024): 1-10.
BibTeX: Download