Increasing the stability of RSA cryptoalgorithm by using genetic optimization of input message.

• Date: 2019
• Main Authors: Pryimak, A. V., Yaremchuk, Yu. E.
• Format: Article
• Language: Ukrainian
• Published: Інститут проблем реєстрації інформації НАН України 2019
• Subjects: cryptostability; RSA; determinism; genetic algorithm
• Online Access: http://drsp.ipri.kiev.ua/article/view/179051
• Journal Title: Data Recording, Storage & Processing

Due to the large amount of data that is transmitted every day via computer networks and stored in cloud environments, network security has become one of the most important aspects of network operation. Cryptography is used to protect information on the network. One of the most popular cryptographic algorithms used in various standards and protocols is RSA, the main drawbacks of which are the cryptographic determinism and vulnerability to attack based on selected ciphertext, so the increase of its crypto stability is still actual issue.The research of the cryptographic algorithm RSA with regard to the possibility of increasing its stability by using genetic algorithm was made. The method of optimization of input message consisting of 8 main steps, with the help of genetic algorithm is proposed. Using the three main properties of the genetic algorithm (selection, crossover and mutation), the input message is randomizing, which as a result of RSA encryption is transformed into a stochastic ciphertext, which is no longer deterministic and weak to attack based on selected ciphertext, and thus increases the cryptostability of this algorithm. The statistical testing of the proposed improvement of the algorithm, using the NIST STS test package, showed a high statistical reliability of this method, as the results of tests were inside 0,9–1 range. Comparison of the results of testing of the original RSA and the modified showed that the original algorithm shows worse performance compared to the proposed its modification. Ten of the fifteen tests showed that the modified RSA algorithm with the built-in proposed method for optimizing the input message has higher rates by 1–3 %, which shows an increase of its cryptostability. Tabl.: 3. Fig.: 5. Refs: 8 titles.