INVESTIGATION OF STATISTICAL CHARACTERISTICS OF NUMERICAL RANDOM SEQUENCE OBTAINED FROM A WEB CAMERA FRAME

Abstract

At the present stage, the generation of random numbers programmatically (pseudo-random sequences) are to a certain extent predictable and susceptible to hacking. Hardware generation in modern conditions is either low-speed or expensive.

The work proposes to use a regular web camera as a source of stochastic chaos. In this case, from a frame of VGA resolution (800 × 600) it is possible to obtain a sequence with a length of 1,440,000 numbers.

The paper presents the results of a study of the statistical characteristics of a sequence of numbers obtained from one frame of a regular webcam, in terms of compliance with one of the requirements of cryptographic information protection: uniform distribution of values. As a result, it was found that even in complete darkness, stochastic thermal processes occurring in the webcam matrix cause an uneven but chaotic distribution of the values ​​of the generated numbers.

It was experimentally established that an absolutely white illuminated surface provides a normal (Gaussian) distribution, while more than 30% of the values ​​are missing, which does not meet the requirements of cryptographic protection and cannot be used for generation. A comparison is made with a similar characteristic of a pseudorandom sequence generated by a software method.

It is shown that by choosing a frame for shooting, it is quite possible to obtain the required level of uniform distribution of sequence elements by value. Determining the level of uniformity of the distribution is carried out quickly using the statistical library of the Java programming language and can be implemented on a regular smartphone, the Android operating system, without the use of cumbersome statistical packages. The results of the study can be used in the design of a hardware random number sequence generator.