Secure Data Encryption Scheme for Cloud Computing

Abhimanyu Jarwal, Kamlesh Kumar, Ajay KumarInternational Journal for Research in Applied Science and Engineering Technology 2023 · DOI

This research paper, published in the International Journal for Research in Applied Science & Engineering Technology (IJRASET), presents a refined approach to securing data in cloud environments. By combining the computational flexibility of Fully Homomorphic Encryption (FHE) with the efficiency of Elliptic Curve Cryptography (ECC), the authors propose a system that promises better security with lower overhead.

TL;DR

The paper proposes a hybrid security scheme for cloud computing that enhances Fully Homomorphic Encryption (FHE) using Elliptic Curve Cryptography (ECC) and One-Time Passwords (OTP). This approach aims to solve the "key management" bottleneck in FHE, resulting in faster data transmission, lower memory usage, and a reduced risk of network breaches.

What Problem Does This Paper Address?

Cloud computing is inherently decentralized, which makes it a prime target for data breaches. Traditionally, security relies on Full Disk Encryption (FDE), which protects data "at rest" but is cumbersome when you need to actually process that data in the cloud.

The "holy grail" of cloud security is Fully Homomorphic Encryption (FHE), which allows a server to perform calculations on encrypted data without ever needing to decrypt it. However, FHE has a major Achilles' heel: complex key management and high computational overhead. The authors address this gap by seeking a "lighter" and more reliable way to manage keys and authenticate users without sacrificing the benefits of FHE.

Methods

The authors introduce a multi-layered security protocol implemented and tested in MATLAB. The key components include:

  • Elliptic Curve Cryptography (ECC): Used for secure channel establishment. ECC is favored over older standards like RSA because it provides equivalent security with much smaller key sizes, making it ideal for mobile devices and high-speed cloud environments.
  • One-Time Password (OTP) Generation: To prevent replay attacks, the system uses a counter-based OTP. Both the cloud server and the user maintain a synchronized "count" that, when combined with a secret key, generates a unique password for every session.
  • HMAC (Hash-based Message Authentication Code): This is used to ensure data integrity. It verifies that the data received by the cloud hasn't been tampered with during transit.
  • The Algorithm Flow: The process follows a strict sequence: User Login $\rightarrow$ ECC Key Generation $\rightarrow$ OTP Authentication $\rightarrow$ HMAC Data Verification $\rightarrow$ FHE Encryption/Decryption.

Datasets

The paper does not utilize a standard public dataset (like those found in machine learning research). Instead, the authors evaluated their scheme using a simulation environment in MATLAB. They compared their proposed "New Scenario" against an "Old Scenario" (likely standard Diffie-Hellman or basic FDE implementations) to measure performance metrics.

Key Insights and Ideas

  • FHE is Function-Centric, not Disk-Centric: The authors emphasize that unlike Full Disk Encryption, which protects the hardware, FHE protects the operation. This is crucial for cloud services where the user doesn't own the physical hardware.
  • Reducing the "Handshake": By using ECC, the proposed scheme requires only two messages to be exchanged to establish a secure channel. This significantly reduces the time spent on authentication compared to traditional methods.
  • Reliability through Key Management: The paper identifies that the primary failure point of FHE is not the encryption itself, but the "reliability of the scheme" regarding how keys are shared. Their ECC-based session key generation aims to make this process more robust.

Key Results

The simulation results showed marked improvements across three main categories:

  • Reduced Latency (Delay): In the "Old Scenario," delay increased significantly as more messages were exchanged. The proposed method maintained a lower, more stable delay (dropping from approximately 35ms to 22ms in the provided charts).
  • Memory Efficiency (Space): The space utilization for the proposed algorithm was lower than the standard Diffie-Hellman approach, dropping from roughly 2500 bytes to 2200 bytes.
  • Security Robustness: The "Probability of Attack" (a metric likely representing the vulnerability window) was reduced from roughly 55% in the existing scenario to 45% with the new protocol.

Limitations and Future Work

While the results are promising, there are a few areas for further exploration:

  • Real-World Scaling: The study was conducted in a controlled MATLAB simulation. How this protocol performs under the massive, unpredictable traffic of a production environment like AWS or Azure remains to be seen.
  • Computational Complexity of FHE: Although the key management is optimized, FHE itself is still computationally expensive. The paper doesn't deeply explore the "noise" growth typically associated with FHE operations.
  • Future Directions: The authors plan to extend this work into access control management, specifically looking at how FHE can be used to manage who has permission to view specific data slices within a cloud database.

Summary updated

Abstract

The cloud computing is the architecture in which no central controller is present due to which various breaches occurred in the network. To secure data transmission from source to destination two type of encryption schemes i.e: fully homomorphism and fully disk encryption are introduced. The fully homomorhic encryption scheme is more security and light as compared to fully disk encryption. In the paper, improvement in the fully homomorphic encryption is proposed using elliptic curve cryptography and OTP generation.
BibTeX
@article{jarwal2023secure,
  author = {Abhimanyu Jarwal and Kamlesh Kumar and Ajay Kumar},
  title = {{Secure Data Encryption Scheme for Cloud Computing}},
  journal = {International Journal for Research in Applied Science and Engineering Technology},
  year = {2023},
  doi = {10.22214/ijraset.2023.51058},
}
Figure