International Journal of Computer Networks and Communications Security

Volume 2, Issue 7, July 2014




Use of Decision Trees and Attributional Rules in Incremental Learning of an Intrusion Detection Model

Pages: 216-224 (9) | [Full Text] PDF (342 KB)
Abdurrahman A. Nasr, Mohamed M. Ezz, Mohamed Z. Abdulmageed
Al-Azhar University, Cairo, Egypt, Faculty of Engineering, Systems and Com. Dept.

Current intrusion detection systems are mostly based on typical data mining techniques. The growing prevalence of new network attacks represents a well-known problem which can impact the availability, confidentiality, and integrity of critical information for both individuals and enterprises. In this paper, we propose a Learnable Model for Anomaly Detection (LMAD), as an ensemble real-time intrusion detection model using incremental supervised machine learning techniques. Such techniques are utilized to detect new attacks. The proposed model is based on making use of two different machine learning techniques, namely, decision trees and attributional rules classifiers. These classifiers comprise an ensemble that provides bagging for decision making. Our experimental results showed that, the model automatically learns new rules from continuous network stream, such that it can efficiently discriminate between anomaly and normal connections, offering the advantage of being deployed on any environment. The model is intensively tested online and its evaluation showed promising results.
Index Terms Decision Trees, AQ, Incremental Classifier, Ensemble, Intrusion Detection.



MadCD: A Mobile Agent Based Distributed Clone Detection Method in Mobile WSNs

Pages: 225-231 (7) | [Full Text] PDF (342 KB)
Department of Electrical and Computer Engineering, Tabriz University, Tabriz, Iran

Sensor nodes used in enemy environments are disposed to capture and Compromise. An adversary may obtain secret information from sensors, such attack is named as clone attack. The clone attack, Replicate nodes and arrange them in the network to launch a variety of other attacks. In recent years, mobile agents have been suggested for effective data broadcasting in sensor networks and a number of researchists use mobile agents as a novel template for distributed purpose to dominate the limitations of sensor nodes. Recently, several solutions are proposed to tackle clone attacks, but they mostly suffer from high overhead. In this paper the Macdc method is proposed to encounter the replication attacks using mobile agent technology in mobile WSNs. The mobile agents are used to aware every node from its trustworthy neighbors, so nodes do not interact with malicious nodes. The analysis and simulation results prove the effectiveness and efficiency of proposed Macdc method which also reduces the comparisons overhead. 
Index Terms Wireless sensor networks, Clone attack, Clone detection, Mobile agent, location.



Image Encryption Using Parallel RSA Algorithm on CUDA

Pages: 23-235 (4) | [Full Text] PDF (806 KB)
Vaibhav Tuteja
School of Information Technology and Engineering, VIT University, Vellore, India

In this paper we discuss Image Encryption and Decryption using RSA Algorithm which was earlier used for text encryption. In today’s era it is a crucial concern that proper encryption decryption should be applied so that unauthorized access can be prevented. We intend to build a general RSA algorithm which can be combined with other image processing techniques to provide new methodologies and better encryption decryption efficiency. One such implementation is by using edge detection method and converting the images to their filtered form. CUDA is a platform for parallel algorithm implementation using CPU with GPU support. The following technique has been implemented on CUDA considering host and device interaction process. Thus, to make the algorithm more efficient we parallelize the algorithm using CUDA block and grid methodology. 
Index Terms Encryption, Decryption, RSA, GPU, Host, Device.



Non-Functional Requirement-Based Service Ranking and Selection

Pages: 236-241 (6) | [Full Text] PDF (261 KB)
Faculty of Sciences Ben M’SIK, Department of mathematics and informatics, Casablanca, Morocco
National School of Computer Science and System Analysis (ENSIAS), Rabat, Morocco

Since more and more applications want to use services that most accurately meet their requirements, we believe that non-functional requirements service selection mechanisms will play an essential role in service-oriented architectures. The integration of these non-functional requirements such as quality, safety, reliability, availability, location, price ... in the process of ranking and selecting services is currently the subject of several research works. Recently published solutions propose using optimization programming techniques. However, the poor scalability of program solving methods restricts their applicability to small-size problems and renders them inappropriate for dynamic applications. In this paper, in addition to a mathematical approach we choose to propose a non functional requirements-based model for selecting services which is, at the same time, generic and abstract. We also consider the notion of user context in order to predict his non functional requirements instead of requesting them, case of figure of most research works. 
Index Terms Services, Ranking, selection, context, model,Non functional requirements.