Real-Time Applications of Deep Learning-Based Steganography in IoT Networks DOI
Harpreet Kaur Channi

Advances in information security, privacy, and ethics book series, Journal Year: 2024, Volume and Issue: unknown, P. 331 - 360

Published: July 12, 2024

The growing amount of resource-constrained IoT networks makes data security and privacy major concerns. In many cases, the most widely used traditional cryptographic techniques fail to be optimal for secure real-time communication in an environment. This work investigates deep learning-based steganography, which aims covertly send confidential information hidden along with seemingly innocuous data. Here, authors examine whether neural can support better hiding as well extraction processes at low overhead, targeting constrained devices. They have analyzed diverse learning architectures—the convolutional networks, autoencoders—improving payload capacitance adding robustness against detection attacks. simulations case studies emphasize applicability these advanced features a wide spectrum applications: smart homes, industrial IoT, health monitoring. Finally, proposed deep-learning-based steganography offers scalable efficient aspects within evolving landscape.

Language: Английский

Real-Time Applications of Deep Learning-Based Steganography in IoT Networks DOI
Harpreet Kaur Channi

Advances in information security, privacy, and ethics book series, Journal Year: 2024, Volume and Issue: unknown, P. 331 - 360

Published: July 12, 2024

The growing amount of resource-constrained IoT networks makes data security and privacy major concerns. In many cases, the most widely used traditional cryptographic techniques fail to be optimal for secure real-time communication in an environment. This work investigates deep learning-based steganography, which aims covertly send confidential information hidden along with seemingly innocuous data. Here, authors examine whether neural can support better hiding as well extraction processes at low overhead, targeting constrained devices. They have analyzed diverse learning architectures—the convolutional networks, autoencoders—improving payload capacitance adding robustness against detection attacks. simulations case studies emphasize applicability these advanced features a wide spectrum applications: smart homes, industrial IoT, health monitoring. Finally, proposed deep-learning-based steganography offers scalable efficient aspects within evolving landscape.

Language: Английский

Citations

0