Scholarlink

<p>Wireless Sensor Networks (WSNs) remain highly vulnerable to malicious nodes due to limited resources, unreliable data, and the absence of secure decentralized validation mechanisms. To address these challenges, this work proposes a Blockchain-based Deep Spatial Pyramid Diffractive Neural Network with Walrus Optimizer (DSPDNNet-WO) for high-accuracy malicious node detection and secure data handling. The framework integrates PoW-based lightweight node authentication, IRMI for robust missing-data correction, and TAEPO for optimal feature selection. The DSPDNNet, enhanced by WO, enables fast and precise classification, while IPFS supports tamper-resistant storage. Experimental evaluation on the WSN-DS dataset shows superior performance with 99.9% accuracy, 99.9% precision, 99.8% recall, 85 Mbps throughput, and 65&#xa0;ms latency. The proposed approach provides a scalable, secure, and computation-efficient solution for intelligent WSN security.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimized storage and cybersecurity in blockchain-assisted wsns through malicious node detection and data enhancement using a deep spatial pyramid diffractive neural network

  • 期刊论文

  • 发表日期：

    2026年1月7日

查看全文
  • R. Mohanapriya,
  • V. Ramu,
  • Krishna Prakash Arunachalam,
  • Rakesh K. Kadu

摘要

Wireless Sensor Networks (WSNs) remain highly vulnerable to malicious nodes due to limited resources, unreliable data, and the absence of secure decentralized validation mechanisms. To address these challenges, this work proposes a Blockchain-based Deep Spatial Pyramid Diffractive Neural Network with Walrus Optimizer (DSPDNNet-WO) for high-accuracy malicious node detection and secure data handling. The framework integrates PoW-based lightweight node authentication, IRMI for robust missing-data correction, and TAEPO for optimal feature selection. The DSPDNNet, enhanced by WO, enables fast and precise classification, while IPFS supports tamper-resistant storage. Experimental evaluation on the WSN-DS dataset shows superior performance with 99.9% accuracy, 99.9% precision, 99.8% recall, 85 Mbps throughput, and 65 ms latency. The proposed approach provides a scalable, secure, and computation-efficient solution for intelligent WSN security.