On the effects of neural network-based output prediction attacks on the design of symmetric key ciphers

Document 2024/1310

On the effects of neural network-based output prediction attacks on the design of symmetric key ciphers

Hayato WatanabeTokai University

Ryoma ItoNational Institute of Information and Communication Technologies

Toshihiro OhigashiTokai University

Abstract

Proving resistance to conventional attacks, e.g., differential, linear, and integral attacks, is essential for designing a secure symmetric key cipher. Recent advances in machine search and deep learning-based methods have made this time-consuming task relatively easy, but concerns persist about expertise requirements and potential forgetfulness. To overcome these concerns, Kimura et al. proposed neural network-based output prediction (NN) attacks, which offer simplicity, generality, and reduced coding errors. NN attacks could be useful for designing secure symmetric key ciphers, especially S-box-based block ciphers. Inspired by their work, we first apply NN attacks to Simon, one of the AND-Rotation-XOR-based block ciphers, and identify structures susceptible to NN attacks and the vulnerabilities detected thereby. Then, we take a closer look at the vulnerable structures. The most vulnerable structure has the weakest diffusion property compared to the others. This fact implies that NN attacks can detect such a property. We then focus on a biased event of the principal function in vulnerable Simon-type ciphers and construct efficient linear approximations caused by such an event. Finally, we use these linear approximations to reveal that the vulnerable structures are more susceptible to a linear key recovery attack than the original attack. We conclude that our analysis can be a solid step toward making NN attacks a useful tool for designing secure symmetric key ciphers.