Posted: 3rd April 2018

We are proud to have a paper accepted at IEEE Symposium on Security and Privacy Symposium (IEEE S&P) 2018. IEEE S&P (aka Oakland)  is one of the most prestigious computer security conferences and has been particularly selective this year, with an acceptance rate of 11.5%.

Our paper 'When Your Fitness Tracker Betrays You: Quantifying the Predictability of Biometric Features Across Contexts' shows how vulnerable different biometrics are against cross-context attacks. The idea of such an attack lies in an imposter collecting biometric data outside the authentication context (e.g. by compromising a fitness tracker or other non-security device) to impersonate the legitimate user. The paper has been co-authored by Oxford's Simon Eberz, Giulio Lovisotto, Andrea Patane, Marta Kwiatkowska and Ivan Martinovic from Oxford and Vincent Lenders from armasuisse. The paper is available online (open access): https://csdl.computer.org/csdl/proceedings/sp/2018/4353/00/435301a853.pdf

The biometric dataset for the paper has been collected with the approval of the Oxford University Central Research Ethics Committee and is available online: https://ora.ox.ac.uk/objects/uuid:0175c157-2c9b-47d0-aa77-febaf07fca71