The Industrial Internet of Things (IIo$T$) is rapidly growing in tandem with security concerns. In this paper, we propose two deep learning models for classifying IIo$T$traffic in binary and multi-class contexts in order to detect intrusions in IIoT networks. To train the models, a recent public dataset is used. The results are very encouraging, with accuracy more than 99%.

Cybersecurity alert prediction is a relevant research field. Based on previous attack steps, the researchers try to know the intention of the attacker and the next attack steps. This paper presents a method that predicts the next future alert depending on previous alerts. The proposed prediction model is inspired from NLP prediction methods and it is a deep-learning method based mainly on LSTM. Our method is compared with different classification methods like SVM, Decision Tree, and Random Forest. The experimental measures show encouraging results, the accuracy reaches 98%.

Authentication is the primary and mandatory process for any Information and Communication Technology (ICT) application to prove the legitimacy of the genuine user. It becomes more important and crucial for public platforms like e-governance platforms. The Government of India is transforming the country into Digital India through various e-governance initiatives based on ICT. For authentication, National e-Authentication Framework (NeAF) was proposed by the Indian government which is a policy framework for authentication. This framework does not provide any technical and unified solution for authentication systems while it is based on centralized verification data. In this paper, we proposed a solution for the authentication which provides the unified authentication solution for the Indian e-governance system with existing infrastructure. This solution also provides the features such as scalability, security, and transparency based on distributed computing and working on multiserver architecture. This solution also fulfills the need of the current Indian government to provide multiple e-governance services through a single smart card.

Cybersecurity attacks are exponentially increasing, making existing detection mechanisms insufficient and enhancing the necessity to design more relevant prediction models and approaches. This issue is still an open research problem since existing attack prediction models are failing to follow the huge amount of attacks and their variety. Recently, machine learning approaches and especially deep learning techniques have received much attention from researchers since their unparalleled high performance in several prediction-based fields. In this context, this paper explores the application of deep learning techniques for predicting cybersecurity attacks. Particularly, it proposes a new LSTM (Long Short-Term Memory), RNN (Recurrent Neural Network), and MLP (Multilayer Perceptron) based models carefully designed to predict the type of attack potentially to hap-pen. The proposed models were validated using a recently available dataset called CTF showing encouraging results especially for the LSTM model with an f-measure greater than 93%.

While the web attacks grow in number and manner, the current web protection methods fail to follow this evolution. This paper introduces a new design of a web application protection method called SPHERES. The main idea behind SPHERES is that it is placed in the application server; it intercepts the decrypted traffic, and checks it against a set of filtering rules specific to the requests. This design allows SPHERES to have the most accurate picture of the exchanged traffic, the websites structures and workflows, the user sessions and their states, and the system states. This accurate picture of the total system allows SPHERES to build a protection sphere around the website and checks several types and levels of protections efficiently. In addition to the detection of known attacks, SPHERES is able to detect zero-day attacks at runtime. The performance study of SPHERES shows that it is much better than two famous existing web protection tools.

Abstract This paper introduces a graph‐based attack description that comes with different analysis methods for alert correlation. The system encompasses an attack scenario detection method, an alert correlation method that recognizes multistep attacks, and graph‐based classification method to extract different types of alerts. The performance analysis shows that the system can correlate a huge number of alerts (more than 442 000 alerts) into a dozens of attack graphs. The attack graph has permitted us to extract several attack properties with high precision. Copyright © 2015 John Wiley & Sons, Ltd.

While the rise of the Internet and the high speed networks made information easier to acquire, faster to exchange and more flexible to share, it also made the cybernetic attacks and crimes easier to perform, more accurate to hit the target victim and more flexible to conceal the crime evidences. Although people are in an unsafe digital environment, they often feel safe. Being aware of this fact and this fiction, the authors draw in this paper a security framework aiming to build real-time security solutions in the very narrow context of high speed networks. This framework is called (f|p) since it is inspired by the elefant self-defense behavior which yields p (22 security tasks for 7 security targets).

This paper presents the design of the lightweight communication interface called RWAPI over the Infini- Band interconnect for clusters of PCs. RWAPI has been developed to provide performance to higher applications on a wide variety of architectures. Since the specifications of the InfiniBand interconnect provides many ways to transfer data, we are discussing some issues regarding the choices between InfiniBand capabilities. We implemented RWAPI using the grid-oriented architecture called GRWA and evaluated the communication performance. We obtained a very low latency and a throughtput very close to the maximum user bandwidth for messages as small as 4 kilo-bytes.