Multi-User Searchable Encryption in the Cloud Cédric Van Rompay, Refik Molva, and Melek Önen ISC 2015 th September 10, 2015 Trondheim, Norway
Outline Searchable Encryption Multi-User Scenarios State of the Art Privacy Issue Challenge Solution Conclusion Cédric Van Rompay 2015-09-10 slide 2
Searchable Encryption Data upload Cédric Van Rompay 2015-09-10 slide 3
Searchable Encryption Data upload Honest-but-curious Cloud Cédric Van Rompay 2015-09-10 slide 4
Searchable Encryption Data upload Honest-but-curious Cloud Encryption for Privacy Cédric Van Rompay 2015-09-10 slide 5
Searchable Encryption Data upload Honest-but-curious Cloud Encryption for Privacy Remote word search queries Cédric Van Rompay 2015-09-10 slide 6
Multi-User Scenarios - PEKS Single Reader Multiple Writers Cédric Van Rompay 2015-09-10 slide 7
Multi-User Scenarios Delegated SE Multiple Readers Single Reader Cédric Van Rompay 2015-09-10 slide 8
Multi-User Scenarios MUSE Multiple Readers Multiple Writers + Search Scalability Cédric Van Rompay 2015-09-10 slide 9
State of the Art for MUSE [Bao2008], [Yang2011], [Popa2014] Scalability through query transformation Cédric Van Rompay 2015-09-10 slide 10
State of the Art for MUSE [Bao2008], [Yang2011], [Popa2014] Scalability through query transformation Data Privacy Cédric Van Rompay 2015-09-10 slide 11
State of the Art for MUSE [Bao2008], [Yang2011], [Popa2014] Scalability through query transformation Data Privacy Search Pattern Privacy (query indistinguishability) Cédric Van Rompay 2015-09-10 slide 12
State of the Art for MUSE [Bao2008], [Yang2011], [Popa2014] Scalability through query transformation Data Privacy Search Pattern Privacy (query indistinguishability) No Access Pattern Privacy (result privacy) Privacy Issue if collusions Cédric Van Rompay 2015-09-10 slide 13
Privacy Issue Cédric Van Rompay 2015-09-10 slide 14
Privacy Issue Access Pattern Similarities Cédric Van Rompay 2015-09-10 slide 15
Privacy Issue Access Pattern Similarities Collusion Data leakage Cédric Van Rompay 2015-09-10 slide 16
Privacy Issue Access Pattern Similarities Collusion Data leakage Privacy Breach: Data Privacy Broken Search Privacy Broken Cédric Van Rompay 2015-09-10 slide 17
Challenge Access Pattern Similarities Collusion Data leakage Privacy Breach: Data Privacy Broken Search Privacy Broken New Adversary Model Cédric Van Rompay 2015-09-10 slide 18
Idea of Solution Private Information Retrieval (PIR) for Privacy Cédric Van Rompay 2015-09-10 slide 19
Idea of Solution Private Information Retrieval (PIR) for Privacy Query Transformation through Proxy Proxy Cédric Van Rompay 2015-09-10 slide 20
Idea of Solution Private Information Retrieval (PIR) for Privacy Query Transformation through Proxy Proxy not Trusted Single Assumption: No collusion between Proxy and Cloud Cédric Van Rompay 2015-09-10 slide 21
Protocol Description Encrypted words upload Cédric Van Rompay 2015-09-10 slide 22
Protocol Description Encrypted words upload Authorization Cédric Van Rompay 2015-09-10 slide 23
Protocol Description Encrypted words upload Authorization Querying Cédric Van Rompay 2015-09-10 slide 24
Protocol Description Encrypted words upload Authorization Querying Query Transformation with bilinear pairings Cédric Van Rompay 2015-09-10 slide 25
Protocol Description Encrypted words upload Authorization Querying Query Transformation with bilinear pairings PIR Querying Cédric Van Rompay 2015-09-10 slide 26
Protocol Description Encrypted words upload Authorization Querying Query Transformation with bilinear pairings PIR Querying (Cheap) PIR Processing Cédric Van Rompay 2015-09-10 slide 27
Protocol Description Encrypted words upload Authorization Querying Query Transformation with bilinear pairings PIR Querying (Cheap) PIR Processing PIR Response Cédric Van Rompay 2015-09-10 slide 28
Achievements Scalable Search New Adv. Model Privacy in new model Access Pattern Privacy Safe against collusion Cédric Van Rompay 2015-09-10 slide 29
Conclusion Multi-User Searchable Encryption with: Extended Adversary Model: Collusions among Cloud (resp. Proxy) and any number of (non authorized) users. Privacy Properties: Data Privacy, Search Pattern Privacy, Access Pattern Privacy Proved external Diffie-Hellman (XDH), PIR Security (IND-CCA2), Random Oracle Model Cédric Van Rompay 2015-09-10 slide 30
Questions? Cédric Van Rompay 2015-09-10 slide 31
References [Bao2008] Bao, F., Deng, R.H., Ding, X., Yang, Y.: Private query on encrypted data in multi-user settings. In: Information Security Practice and Experience, pp. 71 85. Springer (2008) [Yang2011] Yang, Yanjiang, Haibing Lu, and Jian Weng. "Multi-user private keyword search for cloud computing." Cloud Computing Technology and Science (CloudCom), 2011 IEEE Third International Conference on. IEEE, 2011. [Popa2014] Popa, R.A., Zeldovich, N.: Multi-Key Searchable Encryption (2013), http://people.csail.mit.edu/nickolai/papers/popa-multikey-eprint.pdf Cédric Van Rompay 2015-09-10 slide 32
Privacy Definitions Games Cédric Van Rompay 2015-09-10 slide 33
Privacy Definitions Games Cédric Van Rompay 2015-09-10 slide 34
Transition Game 0 1 Cédric Van Rompay 2015-09-10 slide 35
Transition Game 2 3 Cédric Van Rompay 2015-09-10 slide 36