No abstract

With the development of artificial intelligence, deep-learning-based cryptanalysis has been actively studied. There are many cryptanalysis techniques. Among them, cryptanalysis was performed to recover the secret key used for cryptography encryption using known plaintext. In this paper, we propose a cryptanalysis method based on state-of-art deep learning technologies (e.g., residual connections and gated linear units) for lightweight block ciphers (e.g., S-DES, S-AES, and S-SPECK). The number of parameters required for training is significantly reduced by 93.16%, and the average of bit accuracy probability increased by about 5.3% compared with previous the-state-of-art work. In addition, cryptanalysis for S-AES and S-SPECK was possible with up to 12-bit and 6-bit keys, respectively. Through this experiment, we confirmed that the-state-of-art deep-learning-based key recovery techniques for modern cryptography algorithms with the full round and the full key are practically infeasible.

As IoT technology develops, many sensor devices are being used in our life. To protect such sensor data, lightweight block cipher techniques such as SPECK-32 are applied. However, attack techniques for these lightweight ciphers are also being studied. Block ciphers have differential characteristics, which are probabilistically predictable, so deep learning has been utilized to solve this problem. Since Gohr’s work at Crypto2019, many studies on deep-learning-based distinguishers have been conducted. Currently, as quantum computers are developed, quantum neural network technology is developing. Quantum neural networks can also learn and make predictions on data, just like classical neural networks. However, current quantum computers are constrained by many factors (e.g., the scale and execution time of available quantum computers), making it difficult for quantum neural networks to outperform classical neural networks. Quantum computers have higher performance and computational speed than classical computers, but this cannot be achieved in the current quantum computing environment. Nevertheless, it is very important to find areas where quantum neural networks work for technology development in the future. In this paper, we propose the first quantum neural network based distinguisher for the block cipher SPECK-32 in an NISQ. Our quantum neural distinguisher successfully operated for up to 5 rounds even under constrained conditions. As a result of our experiment, the classical neural distinguisher achieved an accuracy of 0.93, but our quantum neural distinguisher achieved an accuracy of 0.53 due to limitations in data, time, and parameters. Due to the constrained environment, it cannot exceed the performance of classical neural networks, but it can operate as a distinguisher because it has obtained an accuracy of 0.51 or higher. In addition, we performed an in-depth analysis of the quantum neural network’s various factors that affect the performance of the quantum neural distinguisher. As a result, it was confirmed that the embedding method, the number of the qubit, and quantum layers, etc., have an effect. It turns out that if a high-capacity network is needed, we have to properly tune properly to take into account the connectivity and complexity of the circuit, not just by adding quantum resources. In the future, if more quantum resources, data, and time become available, it is expected that an approach to achieve better performance can be designed by considering the various factors presented in this paper.

The deep voice detection technology currently being researched causes personal information leakage because the input voice data are stored in the detection server. To overcome this problem, in this paper, we propose a novel system (i.e., DeepDetection) that can detect deep voices and authenticate users without exposing voice data to the server. Voice phishing prevention is achieved in two-way approaches by performing primary verification through deep voice detection and secondary verification of whether the sender is the correct sender through user authentication. Since voice preprocessing is performed on the user local device, voice data are not stored on the detection server. Thus, we can overcome the security vulnerabilities of the existing detection research. We used ASVspoof 2019 and achieved an F1-score of 100% in deep voice detection and an F1 score of 99.05% in user authentication. Additionally, the average EER for user authentication achieved was 0.15. Therefore, this work can be effectively used to prevent deep voice-based phishing.

No abstract