A simple nomogram for fast computing the code entropy for 256-bit codes that artificial neural networks output

Rationale and objectives. The paper aims at the decrease of requirements to a test sample size while estimating the entropy of long codes with dependent digits. Materials and methods. Codes in the Hamming space instead of typical codes are used that is equal to an exponential decrease of a number of...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series 2019-08, Vol.1260 (2), p.22003
Main Authors: Ivanov, A I, Lozhnikov, P S, Bannykh, A G
Format: Article
Language:English
Subjects:
Algorithms
Artificial neural networks
Codes
Complexity
Computation
Entropy
Hamming codes
Nomograms
Normality
Physics
Statistical tests
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Rationale and objectives. The paper aims at the decrease of requirements to a test sample size while estimating the entropy of long codes with dependent digits. Materials and methods. Codes in the Hamming space instead of typical codes are used that is equal to an exponential decrease of a number of controlled code states. Results. Usage of beta distributions instead of normal ones for describing the Hamming distances allows overcoming the restrictions of the previously applied GOST R 52633.3 algorithm. A nomogram generated for fast and reliable computations appeared linear in logarithmic coordinates that makes its usage simple and convenient. Conclusions. The calculation of the entropy of long codes by Shannon method is not possible in practice as it is a task with exponential computational complexity. The usage of the Hamming codes instead of normal ones results in a decrease of the computational complexity to a linear one. The application of beta distributions for describing the Hamming statistics allows overcoming normality test restrictions for initial data. The proposed fast method of computing the entropy is becoming common.
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/1260/2/022003