Reliability-Aware Neural Decoding with Adaptive Multi-Source Information Fusion
Entropy · 13.03.2026 · 28 (3) · 323
DOI: 10.3390/e28030323 ↗Kurz gefasst
Fachartikel in der Zeitschrift Entropy, online erschienen am 13.03.2026. Verfasst von Pengxi Fu und 6 weiteren Autor:innen. Beteiligte Einrichtung: Xi'an Key Laboratory of Intelligent Perception and Autonomous Navigation for Low-Altitude Aircraft, Xijing University, Xi'an 710123, China. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shaanxi Province Key Research and Development Program, Shaanxi Key Laboratory of Integrated and Intelligent Navigation, Foundation of the Shaanxi Province Key Research and Development.
Abstract (Originalsprache)
Modern communication systems increasingly leverage multiple information streams—including channel observations, statistical models, and contextual knowledge—to enhance decoding reliability. However, the varying and often unpredictable quality of these sources poses a critical challenge: rigid combination rules fail when source reliability fluctuates, while manual tuning cannot adapt to dynamic operating conditions. This paper presents a neural decoder architecture that automatically learns to assess and fuse heterogeneous information sources based on their instantaneous reliability. Central to our design is a learnable gating module that dynamically weights information streams, demonstrating emergent Bayesian-like behavior—increasing reliance on statistical models under high uncertainty while transitioning to observation-dominated processing as signal confidence improves. To combat the progressive dilution of auxiliary information in deep architectures, we propose a continuous injection strategy that refreshes auxiliary features at each processing layer through dedicated encoding pathways. The underlying message-passing network adopts a heterogeneous bipartite structure with direction-dependent edge parameterization, respecting the asymmetric computational roles inherent in iterative decoding algorithms. Comprehensive experiments validate that the proposed approach not only improves nominal performance but critically maintains robustness when auxiliary information quality degrades or becomes mismatched with actual conditions.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Xi'an Key Laboratory of Intelligent Perception and Autonomous Navigation for Low-Altitude Aircraft, Xijing University, Xi'an 710123, China.
- School of Computer Science, Northwestern Polytechnical University, Xi'an 710129, China.
ORCID-Profile: Zhen Wang, Jianxin Guo, Zhentao Huang
Förderung
- Shaanxi Province Key Research and Development Program
- Shaanxi Key Laboratory of Integrated and Intelligent Navigation
- Foundation of the Shaanxi Province Key Research and Development
Zitieren
APA
Fu, P., Wang, Z., Guo, J., Zhang, Y., Wang, F., Zhu, R., & Huang, Z. (2026). Reliability-Aware Neural Decoding with Adaptive Multi-Source Information Fusion. Entropy, 28(3), 323. https://doi.org/10.3390/e28030323
Vancouver
Fu P, Wang Z, Guo J, Zhang Y, Wang F, Zhu R, et al. Reliability-Aware Neural Decoding with Adaptive Multi-Source Information Fusion. Entropy. 2026;28(3):323. doi:10.3390/e28030323
BibTeX
@article{Fu2026,
author = {Fu, Pengxi and Wang, Zhen and Guo, Jianxin and Zhang, Yushuai and Wang, Feng and Zhu, Rui and Huang, Zhentao},
title = {Reliability-Aware Neural Decoding with Adaptive Multi-Source Information Fusion},
journal = {Entropy},
year = {2026},
volume = {28},
number = {3},
pages = {323},
doi = {10.3390/e28030323}
}
EUVIMED