"Real World Flow-level Network Traffic Dataset"

"This dataset provides real-world flow-level network traffic collected from the campus WLAN infrastructure of the Indian Institute of Technology (IIT) Jodhpur, India. The data were gathered from multiple locations within the same campus environment, including academic departments, student hostels, and residential areas, capturing diverse network usage patterns across different user communities. Network traffic was exported using NetFlow\/IPFIX and initially captured in PCAPNG format at a centralized monitoring point. The raw packet traces were subsequently processed using a custom Python-based framework to extract individual flow records from the captured packets. During preprocessing, flow-level attributes such as transport protocol, packet counts, byte volumes, and timestamps were reconstructed and converted into structured CSV files for easier analysis. To ensure user privacy and comply with institutional data protection policies, sensitive information such as source and destination IP addresses and other potentially identifying fields was removed or anonymized prior to public release. The resulting dataset provides privacy-preserving flow-level traffic data that can support research in network traffic characterization, anomaly detection, traffic prediction, and AI\/ML-driven network management for next-generation communication systems, including 5G and beyond networks."