Abstract: Wireless sensor networks (WSNs) are network type where sensors are used to collect physical measurements. It has many application areas such as healthcare, weather monitoring and even military applications. Security in this kind of networks is a big concern especially in the applications that required confidentiality and privacy. Therefore, providing a WSN with an intrusion detection system is essential to protect its security from different types of intrusions, cyber-attacks and random faults. Clustering has proven its efficiency in prolong the node as well as the whole WSN lifetime. In this paper we have designed an Intrusion Detection (ID) system based on Stable Election Protocol (SEP) for clustered heterogeneous WSNs. The benefit of using SEP is that it is a heterogeneous-aware protocol to prolong the time interval before the death of the first node. KDD Cup’99 data set is used as the training data and test data. After normalizing our dataset, we trained the system to detect four types of attacks which are Probe, Dos, U2R and R2L, using 18 features out of the 42 features available in KDD Cup'99 dataset. The research used the K-nearest neighbour (KNN) classifier for anomaly detection. The experiments determine K = 5 for best classification and this reveals recognition rate of attacks as 75%. Results are compared with KNN classifier for anomaly detection without using a clustering algorithm.

Manal Abdullah, Ebtesam Alsanee and Nada Alseheymi, “Energy Efficient Cluster-Based Intrusion Detection System for Wireless Sensor Networks” International Journal of Advanced Computer Science and Applications(IJACSA), 5(9), 2014. http://dx.doi.org/10.14569/IJACSA.2014.050902

@article{Abdullah2014,
title = {Energy Efficient Cluster-Based Intrusion Detection System for Wireless Sensor Networks},
journal = {International Journal of Advanced Computer Science and Applications},
doi = {10.14569/IJACSA.2014.050902},
url = {http://dx.doi.org/10.14569/IJACSA.2014.050902},
year = {2014},
publisher = {The Science and Information Organization},
volume = {5},
number = {9},
author = {Manal Abdullah and Ebtesam Alsanee and Nada Alseheymi}
}