Scholarlink

<p>Wireless Sensor Networks (WSNs) integrated with cloud-based infrastructure are increasingly vulnerable to sophisticated botnet assaults, particularly in dynamic Internet of Things (IoT) environments. In order to overcome these obstacles, this study introduces a new framework for intrusion detection based on a Dual-Level Contextual Graph-Informed Neural Network with Starling Murmuration Optimization (DeC-GINN-SMO). The proposed method operates in multiple stages. First, raw traffic data from benchmark datasets (Bot-IoT and N-BaIoT) is securely stored using a Consortium Blockchain-Based Public Integrity Verification (CBPIV) mechanism, which ensures tamper-proof storage and auditability. Pre-processing is then performed using Zero-Shot Text Normalization (ZSTN) to clean and standardize noisy network logs. For feature extraction, the model employs a Geometric Algebra Transformer (GATr) that captures high-dimensional geometric and temporal relationships within network traffic. These refined features are used by the DeC-GINN, which combines graph-based learning and dual-level attention mechanisms to detect malicious activity. To improve detection robustness, model parameters are optimized using a Starling Murmuration Optimization (SMO) algorithm inspired by natural flocking behavior. The proposed architecture integrates advanced normalization, blockchain-based storage, graph-informed learning, and bio-inspired optimization to deliver a scalable and adaptive security solution suitable for real-time botnet attack detection in WSNs.</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

Dual-level contextual graph-informed neural network with starling murmuration optimization for securing cloud-based botnet attack detection in wireless sensor networks

  • 期刊论文

  • 发表日期：

    2025年10月4日

查看全文
  • B. S. Deepa Priya,
  • V. Ramu,
  • Krishna Prakash Arunachalam,
  • D. Santhakumar

摘要

Wireless Sensor Networks (WSNs) integrated with cloud-based infrastructure are increasingly vulnerable to sophisticated botnet assaults, particularly in dynamic Internet of Things (IoT) environments. In order to overcome these obstacles, this study introduces a new framework for intrusion detection based on a Dual-Level Contextual Graph-Informed Neural Network with Starling Murmuration Optimization (DeC-GINN-SMO). The proposed method operates in multiple stages. First, raw traffic data from benchmark datasets (Bot-IoT and N-BaIoT) is securely stored using a Consortium Blockchain-Based Public Integrity Verification (CBPIV) mechanism, which ensures tamper-proof storage and auditability. Pre-processing is then performed using Zero-Shot Text Normalization (ZSTN) to clean and standardize noisy network logs. For feature extraction, the model employs a Geometric Algebra Transformer (GATr) that captures high-dimensional geometric and temporal relationships within network traffic. These refined features are used by the DeC-GINN, which combines graph-based learning and dual-level attention mechanisms to detect malicious activity. To improve detection robustness, model parameters are optimized using a Starling Murmuration Optimization (SMO) algorithm inspired by natural flocking behavior. The proposed architecture integrates advanced normalization, blockchain-based storage, graph-informed learning, and bio-inspired optimization to deliver a scalable and adaptive security solution suitable for real-time botnet attack detection in WSNs.