Student Talks 2016 (abstracts)
Post-quantum Security of the CBC, CFB, OFB, CTR, and XTS Modes of Operation
University of Tartu, Inst. of Computer Science, Estonia
We examine the IND-qCPA security of the wide-spread block cipher modes
of operation CBC, CFB, OFB, CTR, and XTS (i.e., security against quantum adversaries doing queries in superposition). We show that OFB and CTR are secure assuming that the underlying block cipher is a standard secure PRF (a pseudorandom function secure under classical queries). We give counterexamples that show that CBC, CFB, and XTS are not secure under the same assumption. And we give proofs that CBC and CFB mode are secure if we assume a quantum secure PRF (secure under queries in superposition).
Probabilistic Output Analysis
Maja H. Kirkeby
Roskilde University, Denmark
Probabilistic output analysis is a static analysis related to output analysis; where an output analysis reason about the program's output base on the program and a description of the input, the probabilistic output analysis reason about the probabilities of the output given the program and a probability distribution of the input.
I and Mads Rosendahl have published (and accepted for publication) the articles "Probabilistic output analysis by program manipulation" and "Probabilistic Resource Analysis by program manipulation" which both describe some of the theory, implementations of the analysis and the results.
Since static analysis cannot give precise answers, we introduce imprecision using over-approximation or under-approximation. This relates badly with the theory of precise probabilities, and, instead, we use a model of imprecise probabilities. The implementation results will be described and related to the optimal derivable output probabilities described by the Dempster-Schafer theory.
This talk will introduce and motivate the probabilistic output analysis and relate the current results to the imprecise probability model Dempster-Schafer structures.
Building a Type-safe Language for OS Kernel Probes
Universita della Svizzera italiana, Switzerland
Probes are little pieces of executable code injected into the kernel to gather the performance data. To be useful probes must not interfere with the usual system execution and certainly must not crash the system. Since probes are executed in the kernel context one can't rely on usual safety features provided by the OS itself to achieve this. Instead, probe code needs to be proven statically safe. We propose to address this problem by introducing a small dependently typed DSL for writing probes, where the type system is used to provide the required safety guarantees.
This is a joint work with Prof. Nathaniel Nystrom, it was published in PLOS'15 workshop proceedings.
November 26, 2014 19:55 Europe/Helsinki (GMT +02:00)
local organizers, ewscs16(at)cs.ioc.ee
EWSCS'16 page: http://cs.ioc.ee/ewscs/2016/