Multipath Transmission System Using Multi-Layer Network Performance Prediction for Autonomous Driving
IEEE Open Journal of Vehicular Technology · 2026 · 7 · 1724-1739
DOI: 10.1109/ojvt.2026.3701369 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Open Journal of Vehicular Technology, 2026. Verfasst von Takuya Tojo und 8 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Reliable upstream communication is essential for remote monitoring in autonomous driving, yet stable video streaming over commercial mobile networks is difficult under vehicular mobility because of bandwidth fluctuations and coverage holes. This paper proposes a cross-layer framework that integrates multipath transmission, multi-layer network performance prediction, and dynamic video quality control. Network-layer prediction based on radio-quality indicators and application-layer prediction based on historical traffic characteristics are fused to optimize packet scheduling and bitrate control, while latency-based fail-safe mechanisms mitigate inevitable prediction errors. We implement the proposed system on in-vehicle devices and multi-access edge computing (MEC) servers, and evaluate it in long-term public-road field experiments in Japan, including a pre-commercial autonomous bus service. Experimental results show that the proposed network- and application-layer performance predictions achieve high recall (76–98%) in detecting communication degradation events, defined as available bandwidth below 0.5 Mbps or latency above 500 ms. The proposed system reduces the maximum continuous video stall duration from 51.8 s to 2.6 s and the total video stall ratio from 9.6% to 1.5%, while maintaining throughput efficiency.
Zitieren
APA
Tojo, T., Sasaki, M., Okada, M., Azuma, N., Tsubaki, T., Ono, K., Kawano, T., Nakamura, M., & Kawamura, K. (2026). Multipath Transmission System Using Multi-Layer Network Performance Prediction for Autonomous Driving. IEEE Open Journal of Vehicular Technology, 7, 1724-1739. https://doi.org/10.1109/ojvt.2026.3701369
Vancouver
Tojo T, Sasaki M, Okada M, Azuma N, Tsubaki T, Ono K, et al. Multipath Transmission System Using Multi-Layer Network Performance Prediction for Autonomous Driving. IEEE Open Journal of Vehicular Technology. 2026;7:1724-1739. doi:10.1109/ojvt.2026.3701369
BibTeX
@article{Tojo2026,
author = {Tojo, Takuya and Sasaki, Motoharu and Okada, Masaki and Azuma, Nobuhiro and Tsubaki, Takuma and Ono, Kotaro and Kawano, Taichi and Nakamura, Mitsuki and Kawamura, Kenichi},
title = {Multipath Transmission System Using Multi-Layer Network Performance Prediction for Autonomous Driving},
journal = {IEEE Open Journal of Vehicular Technology},
year = {2026},
volume = {7},
pages = {1724--1739},
doi = {10.1109/ojvt.2026.3701369}
}
EUVIMED