Nowadays, computer networks are widely used by different individuals and organizations as a tool for rapid transmission of information. For the Security of users and organizations, various security techniques are used which provide a level of security. Some of these methods include the use of a tunneling protocols and virtual private networks. Although these methods can be used to enhance the security of the network, but in parallel can also be used as a tool for misuse. Tunneling protocols can easily violate security policies of an organization. In this relation, we can cite the security challenges in the network tunneling protocols, as documented in RFC 6169. Therefore, identification of tunneling protocols is critical for traffic management, quality of service and misuse detection. In this literature, payload-based methods and statistical feature-based methods are the two main methods for traffic identification. Due to the advantages and disadvantages of these two approaches, combination of both methods can benefit from both and achieve better results. In this thesis, after presenting a comprehensive justify; LINE-HEIGHT: normal; MARGIN: 0cm 0cm 0pt; unicode-bidi: embed; DIRECTION: ltr" Keywords : Traffic Identification, Tunneling Protocols, Packet Content Analysis, Flow Statistical Features, Hybrid Methods
