author        = {Michael Hutter and Mario Kirschbaum and Thomas Plos and J\"{o}rn-Marc Schmidt},
  title         = {Test Apparatus for Side-Channel Resistance Compliance Testing},
  booktitle     = {Non-Invasive Attack Testing Workshop -- NIAT, Nara, Japan, September 26-27},
  year          = {2011},
  keywords      = {Non-Invasive Attack Setup, Test Methods, Side-Channel Resistance, DPA, SPA, DEMA},
  url           = {http://csrc.nist.gov/news_events/non-invasive-attack-testing-workshop/papers/index.html},
  abstract      = {A way to classify the security level of a cryptographic device is to estimate the effort an adversary has to invest in an attack to be successful. While there are metrics and mathematical models to determine the complexity of attacks on cryptographic algorithms and protocols, estimating the security level of an implementation is more complicated. This is because attacks on the implementation depend on a variety of parameters: the expertise of the adversary, the equipment that is available, the knowledge about the implementation, and the individual information leakage of the device. In this paper, we propose a low cost test apparatus that allows amplifying the side-channel leakage by using a second device for noise cancelation. This technique improves the quality of side-channel measurements even without detailed knowledge and control over the reference device. We evaluated our idea by designing and evaluating three different apparatus each using two cryptographic devices. We achieved a side-channel leakage improvement between 20 and 220% compared to a classical side-channel attack setup using only one device. The number of needed traces is reduced by a facto of 10 which not only minimizes the effort in evaluating the side-channel resistance of countermeasure-enabled devices but also helps in performing effcient attacks in practice.}