An Urban Multi-Operator QoE-Aware Dataset for Cellular Networks in Dense Environments

2025-06-23

AI Summary

Urban cellular networks face insufficient data support for QoE modeling and optimization under high-density deployments, diverse user mobility patterns, and heterogeneous traffic demands. To address this, we construct the first empirically collected, multi-operator QoE-aware dataset tailored to dense urban environments—covering three major operators, multiple mobility modes (walking, BRT, shuttle bus), and representative applications (HTTP browsing, video streaming, FTP). Crucially, the dataset integrates multi-operator configuration parameters, fine-grained mobility trajectories, and user-centric QoE labels derived from key radio metrics (RSRP, RSRQ, SNR), with 132 physical cells geolocated and validated via OpenCellID and on-site surveying. We release a structured CSV dataset comprising 30,925 annotated records, accompanied by preprocessing and visualization scripts. This resource enables machine learning–driven network planning and optimization research for 5G and heterogeneous wireless networks.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Application Domains: Mobility, Driving & Flight
• Search and Optimization: Evaluation and Analysis

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Urban cellular networks face complex performance challenges due to high infrastructure density, varied user mobility, and diverse service demands. While several datasets address network behaviour across different environments, there is a lack of datasets that captures user centric Quality of Experience (QoE), and diverse mobility patterns needed for efficient network planning and optimization solutions, which are important for QoE driven optimizations and mobility management. This study presents a curated dataset of 30,925 labelled records, collected using GNetTrack Pro within a 2 km2 dense urban area, spanning three major commercial network operators. The dataset captures key signal quality parameters (e.g., RSRP, RSRQ, SNR), across multiple real world mobility modes including pedestrian routes, canopy walkways, shuttle buses, and Bus Rapid Transit (BRT) routes. It also includes diverse network traffic scenarios including (1) FTP upload and download, (2) video streaming, and (3) HTTP browsing. A total of 132 physical cell sites were identified and validated through OpenCellID and on-site field inspections, illustrating the high cell density characteristic of 5G and emerging heterogeneous network deployment. The dataset is particularly suited for machine learning applications, such as handover optimization, signal quality prediction, and multi operator performance evaluation. Released in a structured CSV format with accompanying preprocessing and visualization scripts, this dataset offers a reproducible, application ready resource for researchers and practitioners working on urban cellular network planning and optimization.

Problem

• Lack of datasets capturing user-centric QoE and diverse mobility patterns
• Need for efficient urban network planning and QoE-driven optimizations
• Absence of multi-operator performance data in dense urban environments

Innovation

• Multi-operator QoE dataset in dense urban area
• Captures diverse mobility and traffic scenarios
• Structured CSV with preprocessing scripts