Cryptographic Algorithm Identification Using Machine Learning and Massive Processing

Mello, Fernando Machado de; Xexeo, Jose Antonio Moreira

doi:10.1109/tla.2016.7795833

Cited by 5 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, Mishra, et. al [6,7] applied PART, C4.5 to the ciphers identification and the accuracy reached over 80%.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Cryptographic Algorithms Identification based on Deep Learning

Xia¹,

Li²,

Chen³

2022

Artificial Intelligence and Machine Learning

View full text Add to dashboard Cite

The identification of cryptographic algorithms is the premise of cryptanalysis which can help recover the keys effectively. This paper focuses on the construction of cryptographic identification classifiers based on residual neural network and feature engineering. We select 6 algorithms including block ciphers and public keys ciphers for experiments. The results show that the accuracy is generally over 90% for each algorithm. Our work has successfully combined deep learning with cryptanalysis, which is also very meaningful for the development of modern cryptography and pattern recognition.

show abstract

“…Recently, Mishra, et. al [6,7] applied PART, C4.5 to the ciphers identification and the accuracy reached over 80%.…”

Section: Related Workmentioning

confidence: 99%

“…Meanwhile, the block ciphers are encrypted in CBC (Cipher Block Chaining) mode for safety. Compared with [6,7], the conditions of our experiments are far more strict, which make our work more meaningful.…”

Section: Public Keys Ciphersmentioning

confidence: 99%

Cryptographic Algorithms Identification based on Deep Learning

Xia¹,

Li²,

Chen³

2022

Artificial Intelligence and Machine Learning

View full text Add to dashboard Cite

show abstract

“…That is, the average distance between any two nodes in the network, where n is the node size in the network. However, for the actual large-scale network, its connectivity is often not guaranteed, that is, there may not be a path between a pair of nodes [11][12]. In order to avoid the divergence of the calculation results, the harmonic average of the distance between nodes in the network is usually defined as the average path length L:…”

Section: Basic Nature Of Networkmentioning

confidence: 99%

Machine Learning Algorithm based on Spatiotemporal Kriging Model in The Construction of National Stock Market Network

2020

IJETC

View full text Add to dashboard Cite

Kriging in space-time has obvious advantages in analyzing the continuous change of natural characteristic quantity. This method can use the time correlation to improve the analysis accuracy. In the analysis of stock market network construction, Kriging is still not applied. The purpose of this paper is to construct the national stock market network based on the machine learning algorithm of spatiotemporal Kriging model. Firstly, the basic theory of the basic properties of the network is introduced. Secondly, the principle of machine learning algorithm is further studied. After distinguishing the classical Kriging model and the spatiotemporal Kriging model, the data of stock market network is constructed. Some problems related to the interpolation with spatiotemporal Kriging method are further studied, such as the reasons for the improvement of precision, the selection of database capacity in the process of spatiotemporal interpolation, etc. The experimental results show that the spatiotemporal Kriging method is used to calculate the spatiotemporal variation function, Cst(0,0) is estimated to be 0.062, and the spatiotemporal variation function is constructed. The average path of stock network is 2.0476, which is close to 2.6095 of random network. At the same time, the clustering coefficient of stock market network is 7619, which is much larger than that of random network. For any real network, if it satisfies the following two conditions at the same time, it can be called a small world effect.

show abstract

“…(1) Machine learning classifier models are the foundation of the majority of identification schemes. For example, Mello and Xexeo use a single support vector machine classifier identification scheme to identify cryptographic algorithms [12]. Fan and Zhao used multiple classification models of random forest, logistic regression, and support vector machine to identify eight block cipher schemes [13].…”

Section: Introductionmentioning

confidence: 99%

Identification of Block Cipher Algorithms Using Multi-Layer Perception Algorithm

Yuan,

Yu,

Yang

et al. 2023

Preprint

View full text Add to dashboard Cite

Cryptographic algorithm identification is the process of distinguishing or identifying the cryptographic algorithm by analysing the potential information of various features in the ciphertext under the condition of known ciphertext, which is the basis of cryptanalysis work. To solve the worse identification accuracy and stability problem of the single-layer scheme in cryptographic algorithm identification work as the complexity of ciphertext enhances, the interference between ciphertext data and the number of encryption algorithms increases, we propose a cryptographic algorithm identification scheme of block cipher algorithm using deep learning algorithm Multi-Layer Perception (MLP) in this paper. In this scheme, 15 NIST randomness test methods are used to extract ciphertext feature, and 10 meaningful features are selected as input to MLP classifier to make predictions. In the ciphertext-only scenario, five block cipher algorithms, AES, 3DES, Blowfish, CAST, and RC2, are selected for the study of cryptographic algorithm identification, then binary classification and multiclassification identification are carried out. The experimental results demonstrate that, when the size of the ciphertext files and other experimental conditions are the same as for the five conventional machine learning models, the proposed scheme has superior accuracy and stability. Among them, the average identification accuracy of binary classification is 76.5% for ciphertext files of 1kB to 512kB size, which is 35.3%, 37.5%, 34.2%, 38.5%, and 41.6% higher than that of the traditional classical machine learning models SVM, GNB, KNN, RF, and LR respectively. The average identification accuracy of the multiclassification is 36.2%, which is significantly higher than the other five classical machine learning algorithms. When the ciphertext file sizes are different, among the identification rates of the six models fluctuate, the MLP model has the greatest stability and the least amount of influence.

show abstract

Cryptographic Algorithm Identification Using Machine Learning and Massive Processing

Cited by 5 publications

References 9 publications

Cryptographic Algorithms Identification based on Deep Learning

Cryptographic Algorithms Identification based on Deep Learning

Machine Learning Algorithm based on Spatiotemporal Kriging Model in The Construction of National Stock Market Network

Identification of Block Cipher Algorithms Using Multi-Layer Perception Algorithm

Contact Info

Product

Resources

About