Differential Cryptanalysis of TweGIFT-128 Based on Neural Network
Runlian Zhang, Mi Zhang, Jiaxu Yan, Yixing Li, Xiaonian Wu, Lingchen Li · 2021
It is a new trend of cryptographic analysis to realize automatic analysis on cryptographic algorithms by means of deep learning in recent years. TweGIFT-128 algorithm is an instantiation tweak block cipher algorithm for encryption authentication scheme ESTATE, and it was selected as a candidate cipher algorithm in the second round NIST challenge. Aiming at the security of TweGIFT-128 algorithm, the neural network model was used to realize automatic differential analysis in this paper, and the problem of searching differential distinguishers was transformed into a classification problem. Then based on DNN, the neural network differential distinguishers on TweGIFT-128 algorithm were constructed, and the 4–8 round neural network differential distinguishers were trained by using fixed difference. Experimental results show that the obtained 4–7 round TweGIFT-128 differential distinguishers can distinguish ciphertext from random data well, and the method constructing distinguisher based on neural network model is simpler and more general than the traditional methods.