True Random Number Generator ZRANDOM USB - True Random Bits

ZRANDOM USB - True Random Number Generator

We offer a professional, patented hardware True Random Number Generator (TRNG) which can be directly connected to a Personal Computer or Notebook.

Fields of application for the ZRANDOM USB generator:

Generation of code numbers or transaction numbers
Encryption of data (e.g. for communication in the Internet)
Digital signatures
Direct use for Monte-Carlo simulations or generation of seed numbers
Reference for pseudo-random number generators
Numeric solution of mathematical problems
Control/test of gambling machines
Creation of lottery numbers

Number generation is based on a physical process, thermal noise. The advantages of true random numbers compared with "pseudo-random" numbers are obvious:

No periodicities.
Not based on an algorithm.
No predictability of random numbers based on knowledge of preceding sequences.
Certainty that no hidden correlations are present.
Certainty that the equipartition fluctuations are purely stochastic. (Pseudo-random numbers contain systematic, unnatural fluctuations in the equipartition.)

A special sampling technique guarantees that there is no bias towards 1's or 0's. The noise energy spectrum and the sampling rate are matched in such a way that the theoretically remaining bit correlation would not be detectable even for an uninterrupted generation time of billion years. No use of free-running oscillators which increase the risk of serial correlations. The Windows 98/ME/2000/XP generation program contains four online random bit tests specified by FIPS PUB140-1 (see below).

Is there a need for True Random Numbers - produced by a Hardware Generator?

Generation rates

There are two modes with different generation rates:

Physical mode: 70,000 bits/s (USB 2.0 interface)

In the physical mode only random bits which are physically generated by sampling thermal electronic noise are used. There is no use of pseudo-random bits at all. In this mode the generation rate is reduced in order to guarantee a very good quality of the random bits. 100% physical generated random bits are used.

XOR mode: 300,000 bits/s (USB 2.0 interface)

In the XOR mode for every resulting (recorded) random bit one physical random bit and one pseudo-random bit have been used. Since the statistical properties of the resulting bits are due to a very good pseudo-random bit algorithm, the physical random bits can be generated with maximum speed. 50% physical generated random bits are used.

(There may be a lower generation rate when using a USB 1.1 interface.)

Security

In both modes absolutely non-deterministic, i.e. unpredictable random bit sequences are generated. During generation the most important statistical properties of the random bits are checked (even if the FIPS PUB140-1 online tests are disabled). No bits which did not pass the tests are recorded. Enabling the FIPS PUB140-1 online tests further increases the security. Here you will find an example protocol file of the online tests.

PC hardware requirements:

Processor: CPU with 300 MHz or higher (CPU with 1,7 GHz or more recommended)
USB connector (USB 2.0 high speed recommended)
CD ROM drive
VGA graphic display

The generation rates mentioned on this page are achieved only with the recommended PC hardware.

The ZRANDOM USB generator consists of an external device which contains the shielded noise generator,
an USB cable, a CD ROM, and a manual.

Supported operating systems:

Windows XP, Windows 2000, Windows ME, Windows 98

Software for Windows 98/ME/2000/XP:

An API (Application Program Interface) is provided as a DLL. Using this DLL it is feasible to generate 20,000 True Random Bits on demand by a function call. The DLL function call is tested for Delphi. With corresponding commands the DLL may also be used in C++ or Visual Basic programs (no guarantee). Here you will find hints how to call the DLL functions.

Conversion to Integer or Real random numbers Win 95/98/ME/NT/2000/XP - (programm info)

Following amount of random bits can be produced per day:

With a generation rate of 70,000 bits/second: about 721 MByte
With a generation rate of 300,000 bits/second: about 3.1 GByte

Binary random bit files created by the ZRANDOM pass the DIEHARD - TEST developed by George Marsaglia without any problems. DIEHARD contains 15 statistical random bit tests and can be regarded as an "acid test" for every random number generator. A lot of generators, like RAN2 from Numerical Recipes, don't pass this test.

Key words:

true random numbers , true random number generator , hardware random number generator , thermal noise , white noise , truly random numbers , high speed generation , random bit , Monte Carlo simulation , encoding , code , encryption , cryptography , crypto , random digit , PGP , pretty good privacy , Vernam code , true random numbers , random bit tests FIPS PUB140-1 , digital signatures , DLL , API Application Program Interface

Back to start page