Novel Key Generator-Based SqueezeNet Model and Hyperchaotic Map

Hayder Najm, Mohammed Salih Mahdi and Sanaa Mohsin

Data and Metadata, 2025, vol. 4, 743

Abstract: Cybersecurity threats are evolving at a very high rate, thus requiring the use of new methods to enhance the encryption of data and the communication process. In this paper, we propose a new key generation algorithm using the simultaneous use of the SqueezeNet deep learning model and hyperchaotic map to improve the hallmark of cryptographic security. The method employed in the proposed approach is built around the SqueezeNet model, which is lighter and faster in extracting features from the input image, and a hyperchaotic map, which is the main source of dynamic and non-trivial keys. The hyperchaotic map enhances complexity and randomness, securing the new cryptosystem against brute force and statistical attacks, and the key length depends on the number of features in the image. All our experiments prove that the proposed key generator works well in generating long, random, high entropy keys and is highly resistant to all typical cryptographic attacks. The promising profound synergy of deep learning and chaotic systems provides directions for the development of secure and effective methods of cryptography amid the exacerbated cyber threats. The technique was found to meet all the 15 criteria as tested through the NIST statistical test suite.

Date: 2025
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:dbk:datame:v:4:y:2025:i::p:743:id:1056294dm2025743

DOI: 10.56294/dm2025743

Access Statistics for this article

More articles in Data and Metadata from AG Editor
Bibliographic data for series maintained by Javier Gonzalez-Argote ().