DNA‑Enhanced Dual‑Layer Encryption and LSB Steganography for Data Security
Background: This paper proposes a practical dual-layer framework that combines dynamic DNA-indexed block substitution with a DNA-guided bidirectional LSB embedding to protect small secret payloads (text, small images, short audio) within cover images. Methods: A session-unique S-box is generated from a DNA subsequence and a cryptographically secure PRNG, producing per-session substitution and pixel transposition; the hiding stage embeds the resulting ciphertext using forward/backward indices (FIndex/LIndex) driven by nucleotide patterns (A, T, G, C). The system was implemented in C# (.NET 4.6.2) and evaluated on standard and medical image sets (cover sizes 256–1024) at payloads up to 0.25 bpp. Results: Results show strong encryption robustness (NPCR = 99.70%, UACI = 33.2%, entropy ≈ 7.98, adjacent-pixel correlation ≈ 0.015) and high stego imperceptibility (average PSNR ≈ 43.5 dB), while runtime remains practical (≈0.95 s to encrypt a 512×512 image; embedding ≈0.3 s). Comparisons against classical baselines (AES-128, 3DES, standard LSB) demonstrate comparable or improved diffusion and perceptual quality. We also present a security analysis, discuss limitations (computational cost, scalability, steganalysis coverage), and outline prioritized future work, including steganalysis testing, formal S-box analysis, optimization, and public code release to ensure reproducibility. Conclusion: The proposed approach offers a balanced trade-off between confidentiality and concealment for sensitive small payloads in realistic settings.
Authors
- Mohammed Abbas Fadhil Al-Husainy (ORCID: https://orcid.org/0000-0003-3554-3614)
- Ghayth Al-Asad
- Mohammad Bani-Hani
- Ala’eddin Al-Zu’bi
- Sara Albatienh
- Hazem Abuoliem
Publication Details
- Journal
- Israa University Journal for Applied Science
- Published
- 2025-09-22
- DOI
- https://doi.org/10.52865/lkek6825
- Primary Topic
- DNA and Biological Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00