Anycast Based Architecture for Identity Hiding and Location Privacy in Mobile Internet

• Main Authors: Lian-Dong Liu, 劉良棟
• Other Authors: Jyh-Cheng Chen
• Format: Others
• Language: en_US
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/99853169641328765471

碩士 === 國立清華大學 === 資訊工程學系 === 91 === The Privacy Network proposed in this thesis can protect the location privacy of Internet users by constructing an overlaid infrastructure over the Internet. It provides privacy points in each network domain that it is connected. Users who have subscribed to privacy services provided by privacy networks can use SIP (Session Initiation Protocol) to manage their sessions. The privacy network is designed based on the observed attribute in NAT (Network Address Translation) mechanism. That is, an outsider can never know the IP address of an insider and the details of the internal network. We design a network that functions like two NAT servers from end-to-end. For a session built between one user and another user, privacy network stands in between these users. The two out-most privacy points act as the carriers for users. The two out-most privacy points in a privacy network act similar to NAT server and forms a dual-NAT. In this way, it hides the callee''s IP address from the caller and the caller''s IP address and identity from the callee. The proposed techniques could protect the privacy of Internet users. The architecture is scalable as well. Our design is useful for various applications. With a normal IP device, users may expose location and identity information to corresponding entities on the Internet in any session. This may result in DDoS (distributed denial of service) and annoy, unwanted pop-up advertisement messages. Even worse, every time the user''s identity is exposed, it may be resell to other commercial distributors and thus results in more ads or annoying calls. In addition, a malicious user on the Internet can collect the location information of a user with periodical ping packets and form another new category in detective business. By deploying the techniques proposed in this thesis, users could hide their identity and location without worrying unsolicited messages.