PUKYONG

Toward Practical Post-Quantum Cryptography for IoT: Ascon and Macro- Based Optimizations of ML-KEM

Metadata Downloads
Abstract
양자컴퓨팅은 기존 공개키 암호 시스템에 심각한 위협이 되고 있으며, 이에 따라 양자 내성 암호의 도입이 필수적이다. NIST 에서 표준화한 ML-KEM(Module Lattice-based Key Encapsulation Mechanism)은 양자 저항성을 제공하지만, 높은 연산량과 메모리 요구로 인해 자원이 제한된 IoT 환경에서는 실용적인 구현이 어렵다. 본 논문에서는 ML-KEM 을 IoT 환경에 효율적으로 적용하기 위한 두 가지 최적화 기법을 제안한다. 첫째, Keccak 기반 구성 요소를 경량의 Ascon 기반으로 대체하여 메모리 사용량을 줄인다. 둘째, 함수 호출로 인한 오버헤드를 줄이기 위해 모듈러 감소 연산을 매크로 형태로 재구현한다. 이러한 최적화를 통해 ML-KEM 의 지연 시간을 19–24% 단축하고, 메모리 사용량은 24%, 코드 크기는 18.8% 감소시키는 성능 향상을 확인하였다. 또한 보안 검증을 통해 최적화가 요구되는 보안 수준을 유지함을 입증하였다. 본 연구는 ML-KEM 의 IoT 기기 실현 가능성을 높이는 방향을 제시한다.|Quantum computing makes public-key cryptography vulnerable to attacks. In response, NIST standardized post-quantum cryptographic algorithms, among which ML-KEM provides lattice-based key encapsulation. However, ML- KEM is difficult to deploy on IoT devices due to its high memory demands and computational costs. To make ML-KEM practical for IoT devices, this thesis proposes two optimization methods. First, Keccak-based primitives are replaced with lighter Ascon-based alternatives to reduce memory usage. Second, modular reduction operations are implemented as macros instead of functions to eliminate function call overhead. These optimizations improve ML-KEM latency by 19–24%, reduce memory usage by 24%, and decrease code size by 18.8%. Security validation confirms that these changes preserve required security guarantees, enabling practical ML-KEM deployment on IoT devices.
Author(s)
Mooh Ludivine Esther Makafui
Issued Date
2025
Awarded Date
2025-08
Type
Dissertation
Keyword
Cryptography, Cyber Security, Post-quantum Cryptography, IoT, Optimization
Publisher
국립부경대학교 대학원
URI
https://repository.pknu.ac.kr:8443/handle/2021.oak/34396
http://pknu.dcollection.net/common/orgView/200000897971
Affiliation
국립부경대학교 대학원
Department
대학원 정보보호학과
Advisor
Piljoo Choi
Table Of Contents
I. Introduction 1
1.1 Motivation and Problem Statement 1
1.2 Research Objectives and Contributions 2
1.3 Outline of Thesis 3
II. Background and Preliminaries 4
2.1 Notations and Abbreviations 4
2.2 ML-KEM Overview 5
2.3 Symmetric Cryptographic Primitives 9
2.3.1 Symmetric Primitives’ Role 9
2.3.2 Keccak vs Ascon 11
2.4 Polynomial Arithmetic and Modular Reduction 14
III. Proposed Methodology 18
3.1 Keccak to Ascon Primitive Replacement 18
3.1.1 Replacement Strategy and Mapping 19
3.1.2 Ascon Primitive Optimizations 20
3.2 Macro-Based Modular Reduction Optimization 24
3.2.1 Function-Call Overhead Analysis. 24
3.2.2 Performance and Code-Size Models 25
3.2.3 Macro Implementation Strategy 27
3.3 Integration of Optimizations 28
IV. Experimental Results 30
4.1 Experimental Setup 30
4.2 Ascon-Based Optimization Results 31
4.3 Macro-Based Modular Reduction Results 32
4.4 Combined Optimization Results 33
4.5 Validation of result 34
V. Security Validation 35
5.1 Security Foundations and Optimization Impact Analysis 35
5.2 Experimental Validation 36
5.2.1 Ascon-Hash512 Pseudorandomness and Independence 36
5.2.2 Matrix Coefficient Uniformity Preservation 38
5.2.3 Noise Vector Independence Verification 40
5.3 Security Validation conclusion 41
VI. Conclusion 43
References 45
Acknowledgement 48
Degree
Master
Appears in Collections:
대학원 > 정보보호학과
Authorize & License
  • Authorize공개
  • Embargo2026-03-01
Files in This Item:
  • There are no files associated with this item.

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.