Detection of VPN Network Traffic

Abstract

Today, in the world of the internet, the amount of internet activity is growing which generates a huge amount of data. Data has become the most powerful asset. There is a tremendous increase in the need for security and privacy for each and every personnel. There is a need for more cybersecurity and privacy preservation. One of the ways to prevent loss of data and breach in privacy is by using a Virtual Private Network. A Virtual Private Network(VPN) sends the data traffic through an encrypted ‘tunnel’, making it extremely difficult to decipher or intercept. Because of the inadequate security, public networks are ideal ground for hackers. Cybercriminals and internet service providers can eavesdrop such networks, which may result in theft of personal/financial data. When a VPN is not used on public networks, hackers may be able to steal personal data such as credit card information or passwords. When an individual is connected to a VPN, the original identity of the user is hidden. Though VPN is used to tunnel the traffic inorder to prevent data breach, it may also become a source for several attacks such as Ransomware. Most of the time, the perpetrator makes use of a VPN to remain anonymous during the attack. As a result, there is a need for VPN detection to prevent such malicious activities. This paper focuses on building a VPN based on a popular network security protocol and detecting VPN traffic using various Machine Learning models. The results are analyzed and accuracy for VPN detection seems to be better than the existing approaches.