An Improved Clone Node Detection System Based on Enhanced Neural Optimization Strategy Over Wireless Sensor Network Environment

Wireless Sensor Networks (WSNs) are widely used in diverse applications, including environmental monitoring and military surveillance. However, their open and distributed nature makes them vulnerable to clone node attacks, where an adversary replicates legitimate nodes to compromise network integrity. This paper presents an improved clone node detection system based on enhanced neural optimization strategies, utilizing various detection protocols such as Energy-efficient Ring-based Clone Detection (ERCD), Distributed Hash Table (DHT), Randomly Directed Exploration (RDE), Random Multicast (RM), and Randomized, Efficient, and Distributed (RED) methods. The performance of these protocols was evaluated based on detection accuracy, energy consumption, communication overhead, and memory usage across different network sizes. Results showed that the RED method achieved the highest detection accuracy of 99.3% in large networks (1000 nodes), followed closely by ERCD with 96.7%. The DHT and RDE protocols demonstrated lower energy consumption, making them suitable for large-scale, resource-constrained WSNs. The RM method had the lowest detection accuracy (94.3%) and the highest communication overhead. Overall, the proposed neural optimization strategy effectively balances accuracy and resource efficiency, enhancing the robustness of WSNs against clone node attacks. Future work will focus on integrating machine learning models to further optimize detection performance and minimize resource consumption.