Can Leakage Models Be More Efficient? Non-Linear Models in Side Channel Attacks

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the last decade, many side channel attacks have been published in academic literature detailing how to efficiently extract secret keys by mounting various attacks, such as differential or correlation power analysis, on cryptosystems. Among the most efficient and widely utilized leakage models involved in these attacks are the Hamming weight and distance models which give a simple, yet effective, approximation of the power consumption for many real-world systems. These leakage models reflect the number of bits switching, which is assumed proportional to the power consumption. However, the actual power consumption changing in the circuits is unlikely to be directly of that form. We, therefore, propose a non-linear leakage model by mapping the existing leakage model via a transform function, by which the changing power consumption is depicted more precisely, hence the attack efficiency can be improved considerably. This has the advantage of utilising a non-linear power model while retaining the simplicity of the Hamming weight or distance models. A modified attack architecture is then suggested to yield the correct key efficiently in practice. Finally, an empirical comparison of the attack results is presented.
Original languageEnglish
Title of host publication2014 IEEE International Workshop on Information Forensics and Security (WIFS),
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages215-220
Number of pages6
ISBN (Electronic)9781479988822
DOIs
Publication statusPublished - 12 May 2014
EventIEEE International Workshop on Information Forensics and Security (WIFS’14) - Atlanta, United States
Duration: 03 Dec 201405 Dec 2014

Conference

ConferenceIEEE International Workshop on Information Forensics and Security (WIFS’14)
CountryUnited States
CityAtlanta
Period03/12/201405/12/2014

Fingerprint Dive into the research topics of 'Can Leakage Models Be More Efficient? Non-Linear Models in Side Channel Attacks'. Together they form a unique fingerprint.

  • Cite this

    Tian, Q., O'Neill, M., & Hanley, N. (2014). Can Leakage Models Be More Efficient? Non-Linear Models in Side Channel Attacks. In 2014 IEEE International Workshop on Information Forensics and Security (WIFS), (pp. 215-220). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/WIFS.2014.7084330