A Data Generator for 5G SFC Network Provisioning
Brigitte Jaumard, Charles Boudreau, Emil Janulewicz · 2024
While there are many studies discussing the placement of Virtual Network Functions (VNFs), the availability of data sets on which to validate the proposed algorithms is limited. The objective of this study is to address the need for a 5G data generator that can alleviate the lack of access to real data. The proposed 5G data generator aims to provide valuable datasets for testing machine learning algorithms for dynamic provisioning of 5G requests, i.e., service requests with a pair of source and destination nodes, a service function chain (SFC), composed of a strict order of VNFs and quality of service constraints (e.g., bandwidth, latency). Different validation tests are proposed to assess the quality of the generated data sets, so as to ensure that the network provisioning problem possesses an adequate level of difficulty.