Page 111 - Kaleidoscope Academic Conference Proceedings 2024
P. 111

QUANTUM-RESISTANT ENCRYPTION FOR SECURE END-TO-END COMMUNICATION


                                                                  1
                                                                              2
                                                      1
                                   1
                        Sameer, Kant ; Neha Kishor, Jadhav ; Jawar, Singh ; Dilip, Singh ; Anjan Kumar, Singh 3
                     1
                      Department of Electrical Engineering, Indian Institute of Technology Patna, Patna, Bihar, INDIA
              2
              Department of Telecommunications, Government of India Office of Additional DG Telecom, MPLSA-Bhopal, India
                                     3
                                      Ministry of Electronics and Information Technology, India
                              ABSTRACT                        problem asymptotically at least as hard as some common
                                                              worst-case lattice problems, proved relatively easy to use
           Cryptographically Relevant Quantum Computers (CRQCs)  in cryptographic constructions, representing a significant
           are no longer hypothetical;  they present significant  advancement [1].
           challenges to current IT infrastructure by potentially
           breaking existing encryption schemes within minutes.  1.1  Related Work
           This  paper  introduces  an  innovative  and  efficient
           method for achieving quantum-resistant encryption through  The National Institute of Standards and Technology (NIST)
           lattice-based cryptography.  We specifically tackle the  has approved four quantum-resistant algorithms. Notably,
           challenge of encrypting extremely small units of data, such  Kyber [9], a Key Encapsulation Mechanism (KEM) based
           as a single letter or a single-bit message, by constructing  on lattice-based cryptography and the Modular Learning
           a multidimensional lattice.  Our proposed technique  with Errors (MLWE) problem, targets the GapSVP problem
           leverages the Short Vector Problem (SVP) in lattice-based  [1]. Other lattice-based algorithms include Crystal-Dilithium
           cryptography and incorporates the Learning with Errors  [21], SABER [15], and NTRU [20].  Kyber is preferred
           (LWE) methodology for data encryption and decryption. We  for its efficiency in secure communication and resistance
           demonstrate the feasibility and robustness of this approach  to quantum attacks, thanks to its use of Number Theoretic
           using a real-time messaging application that provides  Transforms (NTT) and noise during ciphertext generation,
           quantum-resistant end-to-end encryption. Our work has the  which complicates attacks like BKZ [19], LLL reduction,
           potential for deployment in strategic applications, securing  BKW, and primal attacks [16].
           information from the "harvest now, decrypt later" threat, even  Oded Regev’s work [17] was pivotal in proving the quantum
           in the presence of quantum technologies.           hardness of the Learning With Errors (LWE) problem and
                                                              Shortest Vector Problem (SVP), underpinning the security of
              Keywords - Quantum Cryptography, Lattice-based,  these lattice-based methods. Efforts to update the Signal
                  Learning with errors, Short vector Problem,  protocol [13] and the Messaging Layer Security (MLS)
                  Quantum-Resistant, Public-key encryption    protocol developed by the Internet Engineering Task Force
                                                              (IETF) [13] for end-to-end secure messaging are ongoing but
                         1. INTRODUCTION                      not yet completed.
           In today’s rapidly evolving digital landscape, the security of
                                                              1.2 Our Contribution
           sensitive information faces unprecedented threats, especially
           with the advent of quantum computing.  Traditional  In this work, we introduce a lattice-based encryption
           cryptographic algorithms,  such as the widely used  algorithm designed to resist quantum attacks, inspired by the
           RSA (Rivest-Shamir-Adleman) public key encryption, are  Learning With Errors (LWE) scheme and the Shortest Vector
           encountering formidable challenges.  The emergence of  Problem (SVP). By integrating Oded Regev’s encryption
           quantum computing, with its unparalleled computational  methods with techniques from Kyber, we developed a new
           power, poses a significant risk to the reliability of  ciphertext generation approach. Our algorithm optimizes
           these conventional encryption schemes. Cryptographically  public key storage by generating the key matrix from a seed
           Relevant Quantum Computers (CRQCs) can effortlessly  and two vectors, reducing storage and transmission overhead
           break existing algorithms, thereby undermining data security.  while enhancing security through increased ciphertext noise.
           In response to this urgent need, the scientific community  We have implemented this algorithm in a server-client
           initially  focused  on  lattice  cryptography  because  model for end-to-end quantum-resistant communication.
           cryptographic constructions based on lattice designs  The messaging application we are developing utilizes a
           came with security proofs derived from worst-case  CCA-secure public key encryption (PKE) scheme. Unlike
           lattice problems. Ajtai and Dwork [8] proposed the first  the typical approach of exchanging a symmetric key via
           lattice-based encryption scheme, which was later refined and  a quantum-safe Key Encapsulation Mechanism (KEM) like
           streamlined by introducing the concept of the Learning With  Kyber, and then encrypting the message, our method directly
           Errors (LWE) problem. The LWE problem, an intermediate  encrypts the message using public key cryptography. This




            978-92-61-39091-4/CFP2268P @ITU 2024           – 67 –                                   Kaleidoscope
   106   107   108   109   110   111   112   113   114   115   116