Generation of True Random Numbers based on Radioactive Decay

Ammar Alkassar, Thomas Nicolay, Markus Rohe

Seongmo Park, Byoung Gun Choi, Park Kyung Hwan

Giuseppe Martini, Francesco G. Bruno

Efficient and reliable Random Number Generator (RNG) are increasingly demanded in cryptographic applications, such as on-the-fly data encription/decription and as a general tool for data protection. Main requirement of RNG is a true randomness, which is achievable by means of physical randomness sources only. A True-RNG based on two independent Stochastic Processes (SP) is presented and discussed. As primary randomness sources, two radioactive decay processes are used. A radioactive source has many advantages over others noise sources, since the decay rate of any nuclide does not depend on temperature, chemical composition, physical state, environmental conditions, and aging. In addition, in U ore for example, the various nuclides are, at very good approximation, at equilibrium in the decay chain, so that many stationary independent SP's with the same statistics can be simply obtained by detecting and selectively routing, e.g. by pulse amplitude sorting, gamma rays of different energy emitted by different nuclides in the decay chain. Simulation is performed by a numerical code. The random bit stream obtained is tested for randomness by the NIST test suite, and results are shown. A hardware implementation of the system is also presented.

Zhiliang L. Yuan

Ricardo Aguayo, Geoffrey Simms, Peter B. Siegel

A simple experimental setup to produce true random numbers is described. The experiment involves measuring successive times between decays of a radioactive source. We discuss two different ways of comparing these times: one that generates a string of random zeros and ones, and another that produces numbers derivable from the permutation group. It is shown that the methods used in the experiment produce the same results for any random process and can therefore be used as a test for randomness in other physical systems.