An Intrusion Detection System Based on Multi-Level Clustering for Hierarchical Wireless Sensor Networks.

Area of Science:

• Computer Science
• Network Security
• Wireless Sensor Networks

Background:

• Hierarchical wireless sensor networks (WSNs) are vulnerable to various intrusions.
• Existing intrusion detection systems (IDS) may not effectively address the hierarchical structure of WSNs.
• Detecting intrusions at different levels of the network hierarchy is crucial for robust security.

Purpose of the Study:

• To propose a novel multi-level clustering intrusion detection system (IDS) framework for hierarchical wireless sensor networks (WSNs).
• To enhance the security of WSNs by detecting intrusions at both subordinate node and cluster head levels.
• To analytically evaluate and present optimal parameters for the proposed IDS framework.

Main Methods:

• Development of a multi-level clustering framework for WSNs.
• Implementation of two distinct intrusion detection approaches: downward-IDS (D-IDS) for member nodes and upward-IDS (U-IDS) for cluster heads.
• Analytical calculations to determine optimal parameters for D-IDS (maximum hops) and U-IDS (monitoring group size).

Main Results:

• The proposed framework effectively distinguishes between normal and abnormal behavior at multiple hierarchical levels.
• Analytical calculations provided optimal parameter values for D-IDS and U-IDS, enhancing detection efficiency.
• The framework demonstrates a robust approach to detecting intrusions in hierarchical WSNs.

Conclusions:

• The proposed multi-level clustering IDS framework offers an effective solution for securing hierarchical WSNs.
• The dual approach of D-IDS and U-IDS enhances intrusion detection capabilities by addressing different network roles.
• Optimized parameters derived through analytical calculations improve the performance and efficiency of the proposed IDS.