A deep learning based steganography integration framework for ad-hoc cloud computing data security augmentation using...

Area of Science:

• Computer Science
• Cybersecurity
• Artificial Intelligence

Background:

• Cloud computing adoption accelerated due to the need for remote work, increasing security and privacy concerns.
• Existing steganography methods using deep learning have shown limited success in enhancing data hiding.

Purpose of the Study:

• To develop an improved steganography technique for ad hoc cloud systems using deep learning.
• To enhance the security and privacy of data transmission in cloud environments.

Main Methods:

• Phase 1: Established an "Ad-hoc Cloud System" using V-BOINC.
• Phase 2: Implemented a modified steganography and deep learning model for secure data transmission.
• Integrated data images within colored images for covert transmission.

Main Results:

• The proposed deep steganography approach demonstrated high effectiveness in concealing data and images against various attacks.
• The ad hoc cloud system showed superior performance compared to Amazon EC2.
• The systematic steganography model achieved lower message detection rates.

Conclusions:

• Deep learning-based steganography offers a promising solution for enhancing data security in ad hoc cloud systems.
• The developed method provides a robust and efficient way to protect sensitive information in cloud environments.