Welcome to the resource topic for 2023/1823
PQC-NN: Post-Quantum Cryptography Neural Network
Authors: Abel C. H. ChenAbstract:
In recent years, quantum computers and Shor’s quantum algorithm have been able to effectively solve NP (Non-deterministic Polynomial-time) problems such as prime factorization and discrete logarithm problems, posing a threat to current mainstream asymmetric cryptography, including RSA and Elliptic Curve Cryptography (ECC). As a result, the National Institute of Standards and Technology (NIST) in the United States call for Post-Quantum Cryptography (PQC) methods that include lattice-based cryptography methods, code-based cryptography methods, multivariate cryptography methods, and hash-based cryptography methods for resisting quantum computing attacks. Therefore, this study proposes a PQC neural network (PQC-NN) that maps a code-based PQC method to a neural network structure and enhances the security of ciphertexts with non-linear activation functions, random perturbation of ciphertexts, and uniform distribution of ciphertexts. The main innovations of this study include: (1) constructing a neural network structure that complies with code-based PQC, where the weight sets between the input layer and the ciphertext layer can be used as a public key for encryption, and the weight sets between the ciphertext layer and the output layer can be used as a private key for decryption; (2) adding random perturbations to the ciphertext layer, which can be removed during the decryption phase to restore the original plaintext; (3) constraining the output values of the ciphertext layer to follow a uniform distribution with a significant similarity by adding the cumulative distribution function (CDF) values of the chi-square distribution to the loss function, ensuring that the neural network produces sufficiently uniform distribution for the output values of the ciphertext layer. In practical experiments, this study uses cellular network signals as a case study to demonstrate that encryption and decryption can be performed by the proposed PQC neural network with the uniform distribution of ciphertexts. In the future, the proposed PQC neural network could be applied to various applications.
Feel free to post resources that are related to this paper below.
Example resources include: implementations, explanation materials, talks, slides, links to previous discussions on other websites.
For more information, see the rules for Resource Topics .