Smart Privacy Protection for Big Video Data Storage Based on Hierarchical Edge Computing

Area of Science:

• Computer Science
• Data Storage
• Network Security

Background:

• The Internet of Things (IoT) generates vast amounts of non-scalar data, like videos, overwhelming local storage.
• Cloud computing struggles to support heterogeneous IoT environments, necessitating edge solutions.
• Existing methods lack effective smart privacy protection for large-scale video data in distributed IoT.

Purpose of the Study:

• To design a hierarchical edge computing architecture for smart privacy protection of video data storage.
• To propose a low-complexity, high-secure scheme utilizing multi-access edge computing, cloudlets, and fog computing.
• To address the storage and computational challenges of big video data in IoT.

Main Methods:

• Developed a hierarchical edge computing architecture integrating multi-access edge computing, cloudlets, and fog computing.
• Implemented a video data partitioning strategy: significant bits of key frames on local devices, least significant bits to cloudlets, and non-key frames to the cloud.
• Employed two-layer parallel compressive sensing and 2D logistic-skew tent map encryption for non-key frames.

Main Results:

• The proposed scheme effectively divides video data across different storage facilities (local devices, cloudlets, cloud).
• Smart privacy protection is achieved for big video data storage in hierarchical edge computing environments.
• The scheme avoids significant increases in computation burden and storage pressure.

Conclusions:

• The hierarchical edge computing architecture provides an effective solution for secure and efficient video data storage in IoT.
• The proposed low-complexity, high-secure scheme balances privacy protection with resource constraints.
• This approach is suitable for heterogeneous distributed IoT environments like wireless sensor networks.