You are looking at a specific version 20210531:070131 of this paper. See the latest version.

Paper 2021/718

Generic Side-Channel Assisted Chosen-Ciphertext Attacks on Streamlined NTRU Prime

Prasanna Ravi and Martianus Frederic Ezerman and Shivam Bhasin and Anupam Chattopadhyay and Sujoy Sinha Roy

Abstract

In this work, we propose novel side-channel assisted chosen-ciphertext attacks applicable to IND-CCA secure NTRU-based PKE/KEMs. In particular, we propose two types of chosen-ciphertext attacks on Streamlined NTRU Prime which instantiate respectively, a plaintext-checking oracle and decryption-failure oracle to perform full key recovery. We propose efficient strategies to construct chosen ciphertexts to instantiate the aforementioned oracles to perform full key recovery. We perform experimental validation of our attacks on the optimized implementation of Streamlined NTRU Prime KEM obtained from the pqm4 public library, a testing and benchmarking framework for post quantum cryptographic schemes on the ARM Cortex-M4 microcontroller. We positively confirm that both the PC and DF oracle- based attacks result in full key recovery in a few thousand traces with 100% success rate. Masking serves as a concrete countermeasure against our proposed attacks and thus our work stresses on the need for concrete masking countermeasures for the NTRU-based PKE/KEMs to protect against similar chosen-ciphertext based side-channel attacks.

Metadata
Available format(s)
PDF
Category
Public-key cryptography
Publication info
Preprint. MINOR revision.
Keywords
Lattice Based CryptographyStreamlined NTRU PrimeChosen-Ciphertext AttacksSide-Channel AttacksPlaintext Checking OracleDecryption Failure OracleNIST PQC Standardization process
Contact author(s)
PRASANNA RAVI @ ntu edu sg
History
2021-10-14: last of 3 revisions
2021-05-31: received
See all versions
Short URL
https://ia.cr/2021/718
License
Creative Commons Attribution
CC BY
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.