EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AI-Enhanced AES Encryption for Kurdish Unicode Texts

Ziyad Hazim Abduljabbar

Dasinya Journal for Engineering and Informatics · 08.05.2026 · 2 (2)

DOI: 10.65542/djei.v2i2.39 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Dasinya Journal for Engineering and Informatics, online erschienen am 08.05.2026. Verfasst von Ziyad Hazim Abduljabbar. Die Arbeit nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This research presents an efficient and secure method for generating the AES master key, based on statistical features extracted from the linguistic structure of Middle Kurdish texts. These features include text length, word count, frequency of unique letters, bigram and trigram, and text entropy. The numerical feature vector is fed with a random salt value into a secure hashing algorithm (SHA256) to scale and encode it into a 32-bit intermediate key. This intermediate key is processed by a three-layer FFNN with random weights and bias values to output the AES master key. For optimal performance and security, the AES algorithm with a 256-bit key and the GCM operating mode were used. The encryption system was developed using Python. The initial test was performed on ten Kurdish texts. The measured entropy values for all generated master keys were high, approaching the maximum Central Kurdish alphabet entropy of 4.9542 bits/letter. To evaluate the system on a larger number of texts, a dataset named CKLTCD was created, consisting of 3000 Kurdish texts of varying lengths and domains. Keys were generated for all dataset texts. AES encryption and decryption were applied, yielding decrypted texts identical to the originals. The SHA function and FFNN significantly complicated and obscured the relationship between the original text and the generated key. The generated key became more independent and complex, making its analysis and prediction extremely difficult. The test results of cryptographic validation (NIST SP800-22, avalanche effect analysis, correlation analysis, and key sensitivity tests) and attack model evaluation (KPA, CPA, and brute-force attacks) reflected the success of the proposed encryption system's strength and security. Comparisons showed the method's similarity to standard functions (PBKDF2, Argon2, HKDF), validating this alternative method for generating dynamic keys based on Kurdish linguistic characteristics.

Zitieren

APA

Abduljabbar, Z. H. (2026). AI-Enhanced AES Encryption for Kurdish Unicode Texts. Dasinya Journal for Engineering and Informatics, 2(2). https://doi.org/10.65542/djei.v2i2.39

Vancouver

Abduljabbar ZH. AI-Enhanced AES Encryption for Kurdish Unicode Texts. Dasinya Journal for Engineering and Informatics. 2026;2(2). doi:10.65542/djei.v2i2.39

BibTeX
@article{Abduljabbar2026,
  author = {Abduljabbar, Ziyad Hazim},
  title = {AI-Enhanced AES Encryption for Kurdish Unicode Texts},
  journal = {Dasinya Journal for Engineering and Informatics},
  year = {2026},
  volume = {2},
  number = {2},
  doi = {10.65542/djei.v2i2.39}
}
Kostenlos starten →