Descriptor: Uplink Latency and Radio-KPI Dataset for Multi-Connectivity Evaluation in Commercial 5G Networks (ULR-MC5G)
Commercial fifth-generation (5G) networks are increasingly considered for latency-sensitive applications such as remote monitoring, teleoperation, and autonomous mobility. However, public datasets that jointly expose packet-level latency, packet loss, radio-layer measurements, mobility information, and dual-operator multi-connectivity (MC) traces remain limited. This article describes a processed dataset collected over commercial 5G Non-Standalone networks in Aalborg, Denmark, using a vehicle-mounted dual-modem measurement platform. The dataset contains synchronized per-packet uplink and downlink latency measurements, application-level packet-loss indicators, Global Positioning System (GPS) information, Long-Term Evolution (LTE) and New Radio (NR) radio key performance indicators (KPIs), serving- and neighboring-cell metadata, and uplink transmit-power measurements. The records cover urban, suburban, and rural routes under controlled constant-bit-rate User Datagram Protocol (UDP) traffic, with particular emphasis on rural uplink operation across multiple target data rates. The dataset is intended to support reproducible research on uplink reliability, radio-to-application-layer correlation, packet duplication, link aggregation, interface switching, and adaptive MC mechanisms in operational cellular networks. A companion source-code repository provides scripts for loading, processing, analyzing, and reproducing the main dataset-derived metrics. IEEE SOCIETY/COUNCIL IEEE Communications Society DATA TYPE/LOCATION Time-series; per-packet latency traces; radio KPI telemetry; mobile 5G NSA measurements; Aalborg, Denmark DATA DOI/PID 10.21227/xp31-0x80