Edge-Cloud VPN Traffic Analysis over Cross Platforms

The escalating demand for user privacy and data encryption has significantly boosted the volume of encrypted traffic on the contemporary Internet. Nonetheless, there is a shortage of comprehensive research on the properties of Virtual Private Network traffic. In this paper, we explore Edge-Cloud Virtual Private Network traffic in order to understand usage patterns without deep packet inspection. To achieve this, we use a machine learning based approach to analyze traffic flows, and identify Virtual Private Network traffic captured from five applications that belong to four service categories. Our results demonstrate that we are not only able to identify the VPN traffic, but also able to identify the different applications and therefore service categories across different network locations and computing platforms.