A Secure Fine-Grained Micro-Video Subscribing System in Cloud Computing

Chunpeng Ge, Lu Zhou, Jinyue Xia, Pawel Szalachowski, Chunhua Su

Abstract Micro-video sharing has become a cornerstone of modern digital interaction, facilitated by the proliferation of smart devices and cloud computing. However, this ecosystem faces significant security challenges due to the "honest-but-curious" nature of cloud providers and the hardware constraints of mobile devices. Traditional encryption methods often fail to support efficient sharing or fine-grained access control. This paper introduces a novel Fine-Grained Identity-Based Broadcast Proxy Re-Encryption (ID-BPRE) scheme. By leveraging identity-based encryption and proxy re-encryption, the proposed system allows content creators to securely outsource video distribution to the cloud while maintaining strict control over who can access the content, all without the cloud ever accessing the original plaintext.

The Limitations of Traditional Video Sharing

In conventional video service models, security is often an afterthought or relies on the cloud provider to manage access lists. As illustrated in Figure 1, the typical workflow involves a user uploading video data and an access list to the cloud. The cloud then checks the subscriber's credentials against this list before delivering the video.

Traditional video sharing model where the cloud manages access to plaintext data.

This model presents three primary issues:

  1. Privacy Risks: The cloud has access to the access list and potentially the video content itself.
  2. Scalability: Standard Identity-Based Encryption (IBE) requires the owner to manually re-encrypt content for every new subscriber, which is impractical for viral micro-videos.
  3. Device Constraints: Mobile devices lack the processing power to handle the heavy computational load of repeated encryption and decryption required by older proxy schemes.

The Proposed ID-BPRE Architecture

To address these challenges, the researchers developed the Fine-Grained ID-BPRE system. This architecture shifts the heavy lifting of re-encryption to the cloud while ensuring the cloud remains blind to the underlying data. The system utilizes a "broadcast" approach, allowing a single re-encryption operation to serve an entire group of authorized subscribers.

The proposed fine-grained ID-BPRE system for secure micro-video sharing.

As shown in Figure 6, the workflow is transformed:

  • Encryption: The owner encrypts the video once using their identity and specific attributes.
  • Re-Encryption Key Generation: The owner generates a special re-encryption key ($rk_{ID \to U}$) containing the subscriber IDs and an Access Tree (fine-grained rules like "Premium Member" AND "Sports Group").
  • Proxy Re-Encryption: The cloud uses this key to transform the owner's ciphertext into a new ciphertext that the authorized subscriber set ($U$) can decrypt.

Technical Implementation and Security

The system is built on six core algorithms: Setup, KeyGen, Encryption, Re-Encryption Key Generation, Proxy Re-Encryption, and Decryption. To ensure robustness against active attacks, the authors employed the Fujisaki-Okamoto conversion, elevating the security from simple Chosen-Plaintext Attack (CPA) resistance to the more rigorous Chosen-Ciphertext Attack (CCA) security.

Crucially, the system is designed for collusion resistance. Even if a malicious subscriber attempts to cooperate with the cloud server, they cannot bypass the attribute-based access rules defined in the owner's Access Tree.

Performance and Scalability

The researchers evaluated the system using the Pairing-Based Cryptography (PBC) Library. Testing involved varying the attribute set size (from 30 to 70) and the number of subscribers (from 20 to 60).

Key results include:

  • Constant-Time Decryption: Regardless of the number of subscribers in the broadcast set, the time for an individual subscriber to decrypt a video remained constant at approximately 3 milliseconds. This decoupling of group size from individual performance is a major advancement over previous schemes.
  • Owner Efficiency: While the time to generate re-encryption keys grows linearly with the number of attributes, it remains highly practical for mobile use, taking only about 90ms for 70 attributes.

Conclusion and Future Work

This paper successfully demonstrates a secure, fine-grained, and efficient micro-video sharing system that balances the needs of content creators with the realities of cloud computing and mobile hardware. While the current scheme requires the cloud to know subscriber identities to perform the broadcast re-encryption, future work will focus on "hidden identity" schemes to further enhance subscriber privacy from the cloud provider.

Reference: Ge, C., Zhou, L., Xia, J., Szalachowski, P., & Su, C. (2019). A Secure Fine-Grained Micro-Video Subscribing System in Cloud Computing. IEEE Access, 7, 137266-137278.

Summary updated

Abstract

Micro-video sharing is prevalent nowadays due to the rapid development of personal smart devices, fast growing of network bandwidth and the easy-to-use cloud computing. However, the micro-video sharing system is more vulnerable compared to the conventional video service model because of the vulnerability of the cloud and the hardware computing constraints of personal devices. Ensuring the security and privacy of the micro-videos has become critical for micro-video sharing systems. Recently, a number of encryption schemes have been proposed to protect the security of the micro-videos stored in the cloud. However, these approaches cannot support micro-video sharing functionality. Moreover, the encryption-based method cannot provide the fine-grained access control on encrypted micro-videos. In this paper, we present a notion of fine-grained identity-based broadcast proxy re-encryption to address the issue of fine-grained encrypted micro-video sharing. Further, we present a concrete scheme in the proposed notion and proved its semantic security. Performance and evaluation reveal that our scheme is practical and efficient.
BibTeX
@inproceedings{ge2019secure,
  author = {Chunpeng Ge and Lu Zhou and Jinyue Xia and Pawel Szalachowski and Chunhua Su},
  title = {{A Secure Fine-Grained Micro-Video Subscribing System in Cloud Computing}},
  booktitle = {IEEE Access},
  year = {2019},
  url = {https://openreview.net/forum?id=hoJO5OYISK},
}
Figure