| Mobile social networking has been increasingly popular with the explosive growth of mobile devices. By allowing mobile users to interact with potential friends around the real world, it enables new social interactions as a complement to web-based online social networks. Motived by this feature, many exciting applications have been developed, yet the challenge of privacy protection is also aroused. This dissertation studies the problem of privacy preserving in mobile social networks. We propose different mechanisms for various privacy requirements.;The first algorithm we proposed is a secure friend discovery mechanism based on encounter history in mobile social networks, which mainly focuses on the location privacy. By exploring the fact that sharing encounters indicate common activities and interests, our scheme can help people make friends with likeminded strangers nearby. We provide peer-to-peer confidential communications with the location privacy and encounter privacy being strictly preserved. Unlike most existing works that either rely on a trusted centralized server or existing social relationships, our algorithm is designed in an ad-hoc model with no such limitation. As a result, our design is more suitable and more general for mobile social scenarios.;We also develop an efficient customized privacy preserving mechanism, which not only protects the privacy of users' profile, but also establishes a verifiable secure communication channel between matching users. Besides, the initiator has the freedom to set a customized request profile by choosing the interested attributes and giving each attribute a specific value. Moreover, the request profile's privacy protection level is customized by the initiator according to his/her own privacy requirements. We also consider the collusion attacks among unmatched users. To the best of our knowledge, this is the first work to address such security threat.;Our protocol guarantees only exactly matching users are able to communicate with the initiator securely, while as little as possible information can be obtained by other participants. To increase the matching efficiency, our design adopts the Bloom filter to efficiently exclude most unmatched users. As a result, our design effectively protects the profile privacy and efficiently decreases the computation overhead. Our third work for this dissertation explores fine-grained profile matching by associating a user-specific numerical value with each attribute to indicate the level of interest. And the profile similarity is computed with a secure dot-product. While existing studies are mainly focused on leveraging rich cryptography algorithms to prevent privacy leakage, we consider a novel cooperative framework by mixing some random noise with the private data to preserve privacy. By carefully tuning the amount of information owned by each party, we guarantee that the privacy is effectively preserved while the matching result of two profiles can be cooperatively obtained. After giving an introduction of the basic mechanism, we detail two enhanced mechanisms by taking collusion attack and verifiability into consideration. With no expensive encryption algorithms involved, our methods are more practical for real-world applications. |
