Content area
Full text
Muhammad Zeeshan 1 and Huma Javed 2 and Amna Haider 3 and Aumbareen Khan 4
Academic Editor:Wei Wei
1, Institute of Information Technology, Kohat University of Science and Technology (KUST), Kohat 26000, Pakistan
2, Department of Computer Science, University of Peshawar (UoP), Peshawar 25000, Pakistan
3, Institute of Management Sciences, Kohat University of Science and Technology (KUST), Kohat 26000, Pakistan
4, Department of Information Technology and Management Science, Preston University Kohat Campus, Kohat 26000, Pakistan
Received 10 July 2015; Revised 30 September 2015; Accepted 8 October 2015; 1 November 2015
This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
1. Introduction
Wireless sensor networks (WSNs) have become a popular area of research in recent years due to their huge potential to be used in various applications. WSNs offer different challenges and vast new research area in continuation to the various applications. These networks are very restricted in terms of battery power, resources, and the overheads involved in communication. Such a network is highly vulnerable to attacks [1]. In such a scenario there are more chances to compromise reliability, availability, integrity of the sensor data traffic, and the sensors data traffic [2].
Intrusion Detection System (IDS) is adopted to deal with WSN security vulnerabilities as a second line of defense in the layered approach. Authorization, authentication, and key management are the first line of defense for a secure environment. Algorithm for IDS should be simple using low computation, complexity, memory, energy, and highly specialized type of attack [3]. Shamshiobad et al. [4] have categorized data models into three techniques; supervised, semisupervised, and unsupervised based on the available training data. This approach suffers a major drawback because of its limitations to obtain normal data and model all possible anomalous behaviors. In 1987, Denning [4, 5] proposed the first real-time intrusion detection model to detect anomalies in a computer system by comparing the current user behavior with a normal behavior model. However, this technique lacks learning capabilities and is based on a normal model generated online which does not change over time during detection.
Artificial Immune System (AIS) is an alternative solution having learning capabilities based on...