5G communication delay dataset for cloud-based vehicle planning and control.

Zhang, X; Xiong, L; Zhang, P; Feng, H; Huang, J; Wang, X; Tian, M; Li, Z; Yin, D · 2026 · PubMed Central

DOI: 10.1038/s41597-026-07239-7

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper introduces CICV5G, a publicly available dataset designed to support research on communication delay modeling and delay-aware control for Cloud-based Intelligent Connected Vehicles (CICVs). The study addresses a critical gap in Intelligent Transportation Systems: while CICVs rely on vehicle-to-network-to-vehicle (V2N2V) communication for cooperative planning and control, existing research often relies on idealized or stochastic delay assumptions that fail to capture the complex, time-varying nature of real-world 5G latency. Accurate characterization of this delay is essential because latency directly impacts trajectory tracking, control stability, and safety. The authors aim to provide a large-scale, high-frequency, and context-rich dataset that synchronizes network indicators with vehicle motion data, enabling more realistic evaluation of cloud-based planning and control strategies. The dataset was generated using a standardized 5G communication delay testbed at the Tongji University Intelligent Connected Vehicle Evaluation Base. The experimental setup involved a modified BYD Qin Pro electric vehicle equipped with a 5G Customer Premises Equipment (CPE), an onboard computing unit, and GNSS sensors. The vehicle communicated with a cloud platform via MQTT protocol over both private (n8 band) and public (n78 band) 5G networks. The authors employed a split-split-plot experimental design to collect data across three driving scenarios (urban, arterial, and rural/off-road), two network modes, and multiple velocity levels (0–80 km/h). Over an eight-month period, more than 150,000 synchronized records were collected. Each record includes millisecond-level timestamps for message publication and reception, allowing for precise calculation of round-trip time (RTT), alongside vehicle state data (position, heading, velocity) and network indicators (RSRP, SINR, Cell ID). Technical validation of the dataset reveals distinct delay characteristics based on environment and network type. In stable coverage areas like urban and arterial roads, average V2N2V delays ranged from 16 to 20 ms, with the private network exhibiting lower variance than the public network. In rural zones, the private network showed significant fluctuations and disconnections due to limited coverage, whereas the public network maintained more stable connectivity through inter-base-station handovers. Statistical analysis using maximum-likelihood estimation demonstrated that the Gamma distribution consistently provided the best fit for the delay data across all configurations, characterized by a positively skewed distribution with a long tail. This pattern reflects communication-induced effects such as radio access network scheduling and core network congestion. The significance of this work lies in providing a reproducible foundation for developing communication delay models and delay-aware control algorithms. By offering high-frequency, end-to-end delay data aligned with vehicle context, CICV5G enables researchers to move beyond simplified latency assumptions. The dataset supports the modeling of real-world 5G communication behavior, the prediction and compensation of time-varying latency, and the performance evaluation of vehicle-cloud coordination strategies. This resource is particularly valuable for enhancing the safety and stability of CICVs by allowing for more accurate simulation and design of control systems that account for realistic network constraints.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• teleoperation remote driving