A Dynamic Traffic Generator for Elastic 5G Network Slicing
Junior Momo Ziazet, Brigitte Jaumard, Huy Quang Duong, P. Khoshabi, Emil Janulewicz
Abstract
The machine learning rise has created a need for 5G traffic data, which remains scarce despite numerous studies relying on machine learning models and algorithms. In this article, we introduce a traffic generator for 5G traffic provisioning under different traffic usage scenarios including traffic forecast. The generator relies on open data from the vehicular and pedestrian traffic of the City of Montreal, which is refactored in order to generate different classes of network traffic, with different characteristics associated with typical 5G applications, and then with different traffic patterns and peak hours. The outcome is a valuable tool for researchers and practitioners interested in the performance evaluation of 5G network traffic predictions and elastic resource orchestration. We give an illustration of the traffic dynamics coming out of our generator, thanks to the refactoring of the open data of the City of Montreal. It shows that the traffic generated is a good fit for studies aiming at traffic prediction or proactive network provisioning under different elastic resource orchestration policies.