Double cluster heads model for secure and accurate data fusion in wireless sensor networks

Area of Science:

• Computer Science
• Network Security
• Wireless Sensor Networks

Background:

• Secure and accurate data fusion is critical in wireless sensor networks (WSNs).
• Existing clustering models face challenges in ensuring data integrity and detecting compromised nodes.
• Reputation and trust systems are vital for enhancing security in distributed networks.

Purpose of the Study:

• To propose a novel cluster-based data fusion model, the Double Cluster Heads Model (DCHM), for WSNs.
• To enhance the security and accuracy of data fusion in WSNs.
• To introduce a mechanism for identifying and removing compromised sensor nodes.

Main Methods:

• Developed the Double Cluster Heads Model (DCHM) integrating clustering, reputation, and trust systems.
• Implemented a dual cluster head selection process within each cluster based on trust.
• Utilized a dissimilarity coefficient at the base station to validate fused data and manage cluster head reputation.

Main Results:

• The DCHM demonstrated superior performance in data fusion security compared to traditional methods.
• The proposed model achieved high accuracy in data fusion.
• The integrated trust and feedback system effectively identified and blacklisted compromised sensor nodes.

Conclusions:

• The DCHM offers a robust solution for secure and accurate data fusion in WSNs.
• The dual cluster head approach combined with a trust system significantly improves network security.
• The model provides an effective mechanism for real-time detection and mitigation of compromised nodes.