Scholarlink

<p>The rapid expansion of the Internet of Things (IoT) has significantly increased the number of connected users and devices, allowing them to interact and control their physical environment. However, the open nature and connectivity of IoT, combined with the resource limitations of its devices, make it vulnerable to various security threats. Ensuring the security and reliability of data exchanged in IoT networks is crucial. This research presents a secure intrusion detection framework that combines Blockchain technology and Deep Learning to address these security concerns. The framework incorporates the RSA (Rivest, Shamir, and Adleman) algorithm for encrypting data, uses Self-Adaptive Tasmanian Devil Optimization (SA-TDO) for generating optimal encryption keys, and employs the SHA3-512 algorithm to maintain data integrity. Additionally, a hybrid model combining Convolutional Transformer Networks (CTN) and Long Short-Term Memory (LSTM) layers with an Optimized Deep Neural Network (ODNN) is utilized for detecting intrusions in IoT systems. The Deep Neural Network (DNN) component is further enhanced using the Sparrow Search Algorithm (SSA) to improve threat detection accuracy. The comparative analysis with existing models shows that this framework effectively secures IoT networks by identifying and mitigating potential threats, thereby ensuring the integrity of communications.</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

A blockchain framework for enhancing data security and intrusion detection in the internet of things environment using hybrid deep learning model

  • 期刊论文

  • 发表日期：

    2025年8月19日

查看全文
  • Siddharth Gautam,
  • Amarjit Malhotra,
  • Sanjay Kumar Dhurandher

摘要

The rapid expansion of the Internet of Things (IoT) has significantly increased the number of connected users and devices, allowing them to interact and control their physical environment. However, the open nature and connectivity of IoT, combined with the resource limitations of its devices, make it vulnerable to various security threats. Ensuring the security and reliability of data exchanged in IoT networks is crucial. This research presents a secure intrusion detection framework that combines Blockchain technology and Deep Learning to address these security concerns. The framework incorporates the RSA (Rivest, Shamir, and Adleman) algorithm for encrypting data, uses Self-Adaptive Tasmanian Devil Optimization (SA-TDO) for generating optimal encryption keys, and employs the SHA3-512 algorithm to maintain data integrity. Additionally, a hybrid model combining Convolutional Transformer Networks (CTN) and Long Short-Term Memory (LSTM) layers with an Optimized Deep Neural Network (ODNN) is utilized for detecting intrusions in IoT systems. The Deep Neural Network (DNN) component is further enhanced using the Sparrow Search Algorithm (SSA) to improve threat detection accuracy. The comparative analysis with existing models shows that this framework effectively secures IoT networks by identifying and mitigating potential threats, thereby ensuring the integrity of communications.