Scholarlink

<p>In Industry 4.0, power plants, smart factories, and financial networks were among the important infrastructures where industrial systems were automated and certified using information and communication technology (ICT). Safety flaws in outdated ICT systems have been brought to light by this advancement. As a result, different cyber-security strategies are created and researched to address vulnerabilities and cyber-attacks. The uncertified safety of new cyber-security models makes their deployment in MANET environment. To provide dependable distribution and integrate these innovations into MANET, their stability and effectiveness are essential. Consequently, this research creates a novel SecureNet-Convolutional Neural Networks (SecureNet-CNN) method for detecting Next-Generation cyber-security attacks in the context of the MANET. The automatic detection of cyber-attacks is the main motivation behind the SecureNet-CNN approach. In the prediction method, min–max normalization is the main step of data normalization. Additionally, the optimization method is used to select the hyper-parameters for the DL approach. Through a thorough series of simulated validations, the experimental study of the proposed approach is validated. Comparing the proposed methodology to other currently employed approaches, the performance validation showed a superior accuracy rating of 99.5%, precision (95%), Recall (94%), and F1-Score (95%).</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

SECURENET-CNN: An Intelligent Deep Learning Approach for Mitigating Cyber Threats in Manets

  • 期刊论文

  • 发表日期：

    2026年2月10日

查看全文
  • Rudramani Bhutia,
  • Radha Mothukuri

摘要

In Industry 4.0, power plants, smart factories, and financial networks were among the important infrastructures where industrial systems were automated and certified using information and communication technology (ICT). Safety flaws in outdated ICT systems have been brought to light by this advancement. As a result, different cyber-security strategies are created and researched to address vulnerabilities and cyber-attacks. The uncertified safety of new cyber-security models makes their deployment in MANET environment. To provide dependable distribution and integrate these innovations into MANET, their stability and effectiveness are essential. Consequently, this research creates a novel SecureNet-Convolutional Neural Networks (SecureNet-CNN) method for detecting Next-Generation cyber-security attacks in the context of the MANET. The automatic detection of cyber-attacks is the main motivation behind the SecureNet-CNN approach. In the prediction method, min–max normalization is the main step of data normalization. Additionally, the optimization method is used to select the hyper-parameters for the DL approach. Through a thorough series of simulated validations, the experimental study of the proposed approach is validated. Comparing the proposed methodology to other currently employed approaches, the performance validation showed a superior accuracy rating of 99.5%, precision (95%), Recall (94%), and F1-Score (95%).