Security analysis of real-world random number generators
Crypto RUB » Lehre (German) » Abschlussarbeiten


The availability of good random numbers is a fundamental requirement
for cryptography. Real random number generators such as dev/random in
Linux or the Fortuna construction in Windows consist of two
components: First, an entropy accumulation phase where multiple low
entropy sources (e.g., user key strokes or network traffic) are
accumulated into an entropy pool. Second, an extraction phase that
uses the accumulated entropy from the entropy pool to derive good
randomness. A recent series of works [1,2,3] proposes initial steps
for a practice-oriented security analysis of real-world random number
generators. The goal of the Master thesis is to move the analysis
given in [1,2,3] further towards practice by either extending the
model of [1,2,3], or analyzing real-world random number generators in
the model of [1,2,3].

[1]
[2]
[3]


Ansprechpartner: S. Faust