Wireless Sensor Networks (WSNs) are vulnerable to various kinds of security attacks that can compromise many nodes and therefore the performance of the network may be degraded. Failures to prevent intrusions could also decrease the credibility of security services, e.g., data confidentiality, integrity, and availability. Traditional Intrusion Detection Systems (IDS) suffer from many issues in performance and increased overhead which are considered the main challenge in WSNs. The common architecture of WSN is that nodes are organized into a set of clusters; each one contains a set of nodes with a specialized node called Cluster Head (CH) node which is responsible for managing activities through the cluster and communicating with other nodes and a Base Station (BS). The CH plays a critical role in the case attacking WSN in the two cases; signature-based and anomaly-based. This paper proposes an efficient hybrid IDS to analyze and secure WSN in multiple phases because it combines the best features of two different approaches to achieve better performance. In the proposed approach, BS evaluates and updates attacks information for the entire network which is the main advantage where BS doesn't suffer from any limitations in sensor nodes. Moreover, BS selects CH among other nodes based on power capabilities and computational resources. The efficiency and adaptability of the proposed method have been tested by simulation experiments deployed on JiST/Swans simulation. The experimental results show that the proposed system is efficient with acceptable performance in comparison with other hybrid IDSs. The experimental evaluation also expresses that the proposed technique reduces the communication costs on the cluster head (CH) which improves the lifetime of the entire WSN.

This a preprint and has not been peer reviewed. Data may be preliminary.