×
The submission system is temporarily under maintenance. Please send your manuscripts to
Go to Editorial ManagerNonlinear stream ciphers have become a viable alternative to traditional cryptosystems in response to the growing need for secure communication. These ciphers generate a keystream via feedback mechanisms and nonlinear functions, which are then utilized for encryption. Geffe generator system is one of the most keystream generators. Also, these systems have many benefits, like being fast, flexible, and able to create unpredictable and non-repeating keystreams, these systems are susceptible to cryptanalysis attacks, which have the potential to compromise their security. This paper presents the first study of applying chicken swarm optimization (CSO) algorithm in the field of cryptanalysis based on cipher only attack. The standard CSO algorithm and an adaptive multi points CSO (AMPCSO) algorithm are proposed to cryptanalysis nonlinear stream cipher based on Geffe keystream generator. Firstly, the traditional CSO is used to reveal the secret initial values of the Geffe generator. Secondly, an adaptive multi points chicken swarm optimization (AMPCSO) has been proposed to enhance the traditional CSO algorithm to attack Geffe generator systems. The AMPCSO is a new idea to advance the CSO search abilities and improve the foraging behavior of hens and chicks by allowing hens to be influenced by other individuals within the same or different groups and affected by the best individual in the population and enable chicks to learn from four reference points rather than learn from their respective mothers only. Lastly, a new criterion is used to estimate the value of fitness by utilizing a multi-objective fitness function (MOFF), which is grounded on Pareto dominance. The experimental results showed that the CSO and AMPCSO are very effective tools in terms of accuracy, information required, and CPU times when applied to the analysis of nonlinear stream cipher. The AMPCSO required a few characters from ciphertext to attack systems with total LFSRs length up to 59 bits with an appropriate CPU time.
Deep falsification of multimedia content, especially videos and photos, threatens social cohesion (e.g., rumour propagation, extortion, and truth distortion) and must not be ignored. In some cases, this issue requires effective detection solutions. Most studies suggest that convolutional neural networks (CNNs) may not be able to extract complex features like those used in deepfake production. Thus, hybrid approaches that can capture complex features and act as powerful descriptors for binary classification are needed to separate bogus from true content. In this paper, a hybrid algorithm is developed to combine gated recurrent units (GRU) and CNN. The proposed model aims to improve the extraction of complex features by simultaneously capturing instantaneous and spatial features. This approach permits the extraction of implicit features that are vital to the final classification process, especially when dealing with a sequential series within video content. Finally, a dense neural network is used to classify these features. Practically, two data sets were used to train the proposed model: the FaceForensics++ (FF++) and DeepFake Detection Challenge (DFDC) datasets. The evaluation results of the proposed model on the FF++ dataset for the Area Under the Curve (AUC) and F1-score metrics reached 0.88% and 0.85%, respectively. While DFDC is 0.95% and 0.86% for the same metrics, respectively.