Denys Ziniar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Denys Ziniar has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
Arsenii Tverdokhlib
2 shared publications- Artem Antonenko
National University of Life and Environmental Sciences of Ukraine
2 shared publications - Юрій Валентинович Мішкур
Державний університет інформаційно-комунікаційних технологій
2 shared publications - Andrii Balvak
State University of Information and Communication Technologies
1 shared publication - Danyil Solskyi
Державний університет інформаційних та комунікаційних технологій
1 shared publication - Myroslav Buriak
State University of Information and Communication Technologies
1 shared publication
Latest Publications
- Open Access
- 28 November 2024
- Published by Khmelnytskyi National University in Measuring and computing devices in technological processes
- No. 4,p. 394-400
ASPECTS OF APPLICATION OF NEURON NETWORK IN CRYPTOGRAPHY
The article analyzes the possibilities of using neural networks in cryptography to increase the security of encryption keys. The authors focus on growing cyber threats and the importance of implementing modern technologies for information protection. The main goal of the research is to evaluate the effectiveness of the neural network in the exchange of encryption keys, based on the achievements in the field of neural cryptography, as well as the development of new methods of protection against cyber threats. The authors created a neural model that is based on the concept of a parity tree and is used for the exchange of encryption keys. At the preparatory stage, a detailed analysis of existing models of neural networks was performed to assess their compliance with the main goal of the project. Using knowledge from similar studies, the authors developed a special neural model in the Python programming language that implements the theoretical foundations. The next stage included the creation of a test environment that allowed for thorough evaluations, guaranteeing the stability and reliability of the neural network in various conditions. The proposed neural network model can become a secure alternative to the traditional Diffie-Hellman key exchange method. In addition, its expected resistance to quantum decryption takes an important step in strengthening cryptographic protocols against new threats in the age of quantum computing. The model shows high efficiency even in simple configurations. The ability of neural networks to quickly adapt to new threats is especially emphasized, which is critically important for ensuring security in conditions of change. The study also shows that the depth of synaptic connections in a neural network makes it much more difficult for attackers to break the key, reducing the chances of success. The findings highlight the wide range of potential applications of neural networks in areas such as cybersecurity, telecommunications, and financial forecasting. Despite certain difficulties associated with algorithms and high requirements for computing resources, neural networks show significant potential for improving cryptographic systems.
- 28 November 2024
- Published by Khmelnytskyi National University in Measuring and computing devices in technological processes
- No. 4,p. 394-400
Open AccessASPECTS OF APPLICATION OF NEURON NETWORK IN CRYPTOGRAPHY
The article analyzes the possibilities of using neural networks in cryptography to increase the security of encryption keys. The authors focus on growing cyber threats and the importance of implementing modern technologies for information protection. The main goal of the research is to evaluate the effectiveness of the neural network in the exchange of encryption keys, based on the achievements in the field of neural cryptography, as well as the development of new methods of protection against cyber threats. The authors created a neural model that is based on the concept of a parity tree and is used for the exchange of encryption keys. At the preparatory stage, a detailed analysis of existing models of neural networks was performed to assess their compliance with the main goal of the project. Using knowledge from similar studies, the authors developed a special neural model in the Python programming language that implements the theoretical foundations. The next stage included the creation of a test environment that allowed for thorough evaluations, guaranteeing the stability and reliability of the neural network in various conditions. The proposed neural network model can become a secure alternative to the traditional Diffie-Hellman key exchange method. In addition, its expected resistance to quantum decryption takes an important step in strengthening cryptographic protocols against new threats in the age of quantum computing. The model shows high efficiency even in simple configurations. The ability of neural networks to quickly adapt to new threats is especially emphasized, which is critically important for ensuring security in conditions of change. The study also shows that the depth of synaptic connections in a neural network makes it much more difficult for attackers to break the key, reducing the chances of success. The findings highlight the wide range of potential applications of neural networks in areas such as cybersecurity, telecommunications, and financial forecasting. Despite certain difficulties associated with algorithms and high requirements for computing resources, neural networks show significant potential for improving cryptographic systems.
- Open Access
- 30 May 2022
- Published by Ukrainian Research and Design Institute of Building Materials and Products in European Science
- No. sge40-02,p. 63-86
OPTIMIZATION OF NETWORK INFRASTRUCTURE TO ENSURES RESILIENCE, SECURITY AND SCALABILITY
- 30 May 2022
- Published by Ukrainian Research and Design Institute of Building Materials and Products in European Science
- No. sge40-02,p. 63-86
Open AccessOPTIMIZATION OF NETWORK INFRASTRUCTURE TO ENSURES RESILIENCE, SECURITY AND SCALABILITY