MUTE-AES: a multiprocessor architecture to prevent power analysis based side channel attack of the AES algorithm

  • Authors:
  • Jude Angelo Ambrose;Sri Parameswaran;Aleksandar Ignjatovic

  • Affiliations:
  • University of New South Wales, Sydney, Australia;University of New South Wales, Sydney, Australia;University of New South Wales, Sydney, Australia

  • Venue:
  • Proceedings of the 2008 IEEE/ACM International Conference on Computer-Aided Design
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Side channel attack based upon the analysis of power traces is an effective way of obtaining the encryption key from secure processors. Power traces can be used to detect bitflips which betray the secure key. Balancing the bitflips with opposite bitflips have been proposed, by the use of opposite logic. This is an expensive solution, where the balancing processor continues to balance even when encryption is not carried out in the processor. We propose, for the first time, a multiprocessor algorithmic balancing technique to prevent power analysis of a processor executing an AES cryptographic program, a popular encryption standard for embedded systems. Our technique uses a dual processor architecture where two processors execute the same program in parallel, but with complementary intermediate data, thus balancing the bitflips. The second processor works in conjunction with the first processor for balancing only when the AES encryption is performed, and both processors carry out independent tasks when no encryption is being performed. Accessing the encryption key or the input data by the first processor begins the obfuscation by the second processor. To stop the encryption by the second processor, we use a novel signature detection technique, which detects the end of the encryption automatically. The multiprocessor balancing approach (MUTEAES) proposed here reduces performance by 0.42% and increases the size of the hardware by 2X (though reduces to 0.1% when no encryption is being performed). We show that Differential Power Analysis (DPA) fails when our technique is applied to AES. We further illustrate, that by the use of this balancing strategy, the adversary is left with noise from the power profile with little useful information.