LSTM-based generation of cellular network traffic

Abstract

Domain-wide recognized by their high value in human activity and network monitoring studies, cellular network traffic (i.e., Charging Data Records, named CDRs), however, present accessibility and usability issues, restricting their exploitation and research reproducibility. This paper tackles such challenges by modeling CDRs that fulfill real-world data attributes. Our designed framework, named Zen leverages LSTM to realistically model network users’ traffic behavior through a 4-stage generative pipeline. Results show that Zen’s models accurately capture individual and global distributions of a fully anonymized real-world traffic CDRs dataset. Finally, we validate Zen CDRs ability of reproducing daily cellular behaviors of the urban population and its usefulness in practical networking applications such as Radio Access Network’s power savings, and anomaly detection as compared to real-world CDRs.