Slide 1Slide 2Slide 3Slide 4Slide 5Slide 6Slide 7Slide 8Slide 9Slide 10Slide 11Slide 12Slide 13Slide 14Towards measuring anonymityC. Diaz, S. Stefaan, J. Claessens, B. PreneelPET’02Presented by B. Choi, cs6461Computer ScienceMichigan TechIntroduction●Applications–Electronic payment, electronic voting, electronic auctions, email and web browsing●No well established framework to assess the degree of anonymity by 2001?–Some proposals: anonymity set size–Shannon’s theory of entrophy in 1948 could be a vehicle to approach the problemSystem model●Anonymity set fixed ans static●Same number of sent messaged by all senders●Senders behavior modeled as a Poisson process●Mix-net anonymity system●Attack model–The degree of anonymity depends on the probabilities that the users have sent a particular message: these probabilities are assigned by the attackerProposed measurement model●Anonymity is the state of being not identifiable within a set of subjects, the anonymity set.●Consider only sender anonymity●Degree of anonymity provided by the system–The quality of the system–Depends on the distribution of probabilities and not on the size of the anonymity setProposed measurement modelExampleExampleCase study: CrowdsCrowdsCrowdsOnion routingOnion routingOnion routingOpen problems●How to find the probability distribution in real situations?●Understanding of real attacks?–We still don’t know how hard or easy to monitor part of or entire of an anonymity system●Core router-based●P2P-based●Degree of anonymity is then relative to attackers–Any standardized absolute degree
View Full Document