Scholarlink

<p>The high growth rate of 6G based IoT ecosystems intensifies the intricacy and the size of cyber threats that require smart, adaptive, and trust-sensitive intrusion detection. The proposed work suggests a new collaborative intrusion detection architecture DySN-GAtN-NRO, combining Maximum-Entropy Regularized Decision Transformer (MaxERDT)-based learning features, Reduced Noise Cluster-Based Synthetic Minority Over-sampling Technique (RNC-SMOTE) to balance data with noise and dynamic Slimmable Global Attention Network (DySN-GAtN) that can reduce model width, as well as the identification of global dependencies. Optimization of Newton -Raphson type (NRO) is also used to make the model more refined and to make the model converge faster and more accurately. To facilitate safe and decentralized exchange of intelligence, the detections are confirmed using Proof of Trust and Expertise (PoTaE) blockchain mechanism. It shows that experimental analysis has a major enhancement in detection accuracy, robustness, scalability and collaborative propagation of trust. The suggested system creates an effective end-to-end defence mechanism designed to work in next-generation 6G-IoT.</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

Collaborative intrusion detection in 6G-enabled IoT networks using Newton Raphson Optimized Dynamic Slimmable Global Attention Network with Blockchain Approach

  • 期刊论文

  • 发表日期：

    2025年12月27日

查看全文
  • G. B. S. R. Naidu,
  • Elangovan Muniyandy,
  • Krishna Prakash Arunachalam,
  • V. S. Narayana Tinnaluri

摘要

The high growth rate of 6G based IoT ecosystems intensifies the intricacy and the size of cyber threats that require smart, adaptive, and trust-sensitive intrusion detection. The proposed work suggests a new collaborative intrusion detection architecture DySN-GAtN-NRO, combining Maximum-Entropy Regularized Decision Transformer (MaxERDT)-based learning features, Reduced Noise Cluster-Based Synthetic Minority Over-sampling Technique (RNC-SMOTE) to balance data with noise and dynamic Slimmable Global Attention Network (DySN-GAtN) that can reduce model width, as well as the identification of global dependencies. Optimization of Newton -Raphson type (NRO) is also used to make the model more refined and to make the model converge faster and more accurately. To facilitate safe and decentralized exchange of intelligence, the detections are confirmed using Proof of Trust and Expertise (PoTaE) blockchain mechanism. It shows that experimental analysis has a major enhancement in detection accuracy, robustness, scalability and collaborative propagation of trust. The suggested system creates an effective end-to-end defence mechanism designed to work in next-generation 6G-IoT.