透過您的圖書館登入
IP:3.142.121.153

摘要


SSL/TLS (Secure Sockets Layer/Transport Layer Security) protocols have been widely used in data transmission to protect the security and integrity of the data. However, due to the encryption of SSL/TLS, the application data over transmitted packets are invisible and difficult to be distinguished by traditional port-based and DPI(Deep Packet Inspection) ways. Though the method based on statistical features can overcome shortages of the above two ways, it is still hard to achieve fine-grained identification. In this paper, we proposed a solution to extract fingerprint information and then identify the types of flows during handshake phase to avoid inspecting the encrypted data and privacy violation. Besides, two hash tables are built to help fast identify the packets with the same APP ID in the same or among different conversations. Finally, 300 flows are captured and the experiment results show the method is accurate and efficient.

關鍵字

Fine-Grained Identification SSL/TLS

延伸閱讀