Outsourcing Verifiable Distributed Oblivious Polynomial Evaluation from Threshold Cryptography

Hamidi, Amirreza, and Ghodosi, Hossein (2023) Outsourcing Verifiable Distributed Oblivious Polynomial Evaluation from Threshold Cryptography. In: Lecture Notes in Computer Science (14252) pp. 235-246. From: ICICS 2023: 25th International Conference on Information and Communications Security, 18-20 November 2023, Tianjin, China.

[img] PDF (Published Version) - Published Version
Restricted to Repository staff only

View at Publisher Website: https://doi.org/10.1007/978-981-99-7356-...
 
1


Abstract

Distributed oblivious polynomial evaluation (DOPE) is a variant of two-party computation where a sender party P1 has a polynomial f(x) of degree k and the receiver party P2 holds an input α. They conduct a secure computation with a number of t distributed cloud servers such that P2 obtains the correct output f(α) while the privacy of the inputs is preserved. This system is the building block of many cryptographic models and machine learning algorithms.

We propose a lightweight DOPE scheme with two separate phases: setup and computation, which means that the setup phase can be executed at any time before the actual computation phase. The number of the servers (t) does not depend on the polynomial degree (k), and the main expensive computation is securely outsourced to the cloud servers using the idea of threshold cryptography. As a result, any normal user with low computational power devices (e.g., mobile, laptop, etc.) would be able to evaluate and verify the output over a large field while the security conditions are preserved. Our protocol maintains the security against a static active adversary corrupting a coalition of up to t − 1 servers and the opposed party. The main two parties commit to their inputs using non-interactive zero-knowledge proof techniques. The communication complexity is linear and bounded to O(t) field elements which means that, unlike the previous studies in this field, it does not depend on the polynomial degree k.

Item ID: 81368
Item Type: Conference Item (Research - E1)
ISBN: 978-981-99-7356-9
Copyright Information: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.
Date Deposited: 13 Dec 2023 22:42
FoR Codes: 46 INFORMATION AND COMPUTING SCIENCES > 4604 Cybersecurity and privacy > 460401 Cryptography @ 40%
46 INFORMATION AND COMPUTING SCIENCES > 4604 Cybersecurity and privacy > 460403 Data security and protection @ 30%
46 INFORMATION AND COMPUTING SCIENCES > 4604 Cybersecurity and privacy > 460499 Cybersecurity and privacy not elsewhere classified @ 30%
SEO Codes: 22 INFORMATION AND COMMUNICATION SERVICES > 2201 Communication technologies, systems and services > 220199 Communication technologies, systems and services not elsewhere classified @ 50%
22 INFORMATION AND COMMUNICATION SERVICES > 2299 Other information and communication services > 229999 Other information and communication services not elsewhere classified @ 50%
Downloads: Total: 1
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page