Online Entropy Estimation for Non-Binary Sources and Applications on iPhone
Year: 2011
Type: report
Abstract: The design of a random number generator is a challenging task on systems in changing environment such as smartphones. Finding reliable and high-throughput sources of entropy is difficult. This paper proposes an online entropy estimation algorithm to test the quality of an entropy source when nothing is known \textit{a priori} on the source statistics. Our estimator can be executed at a low cost and is adapted for any type of sources. It extends the results of Bucci and Luzzi to non-binary sources and introduces a parameter that allows to trade time and memory for a better estimate. Our estimator is then applied to several sources available on an iPhone and compare to the state of the art.
Source: HAL
Authors: Cédric Lauradoux, Julien Ponge, Andrea Roeck
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2211416055
