Scholarlink

<p>Vehicular communications are applied in different security and business applications in the digital world for the user’s advantage. During this, vehicle-to-vehicle (V2V) communication allows clients to exchange necessary information with neighbouring vehicles directly. Generally, vehicles move quickly on the highway instead of the crossing road environment, and therefore, a strong method is needed to communicate securely and efficiently. Vehicular Ad Hoc Networks (VANETs) have several distinct features, comprising fast movable nodes, scattered networks, often changing topologies and self-organisation. A malicious attack changes the packet information in the fabrication attack, causing jamming and higher delays in the vehicular network. In this study, a Quantum-Resistant Hybrid Encryption for Secure Vehicle-to-Vehicle Communication Using Deep Representation Learning (QRHEV2V-DRL) method is proposed. The primary aim of the QRHEV2V-DRL method is to demonstrate significant potential in the development of V2V communication networks to ensure long-term data security in the face of evolving cyber threats. In this V2V communication stage, the QRHEV2V-DRL method performs cluster formation using the Kernel Fuzzy C-Means (KFCM) model to partition the communication network into distinct clusters based on similarities in in-vehicle communication patterns. In this attack detection stage, the QRHEV2V-DRL approach utilises min–max normalisation to scale the features within a particular range. Afterwards, the stacked sparse autoencoder (SSAE) model is used for attack classification. In the encryption stage, the quantum-resistant hybrid encryption (QRHE) model is employed to transmit the data in the CH securely to the cloud. The comparison analysis of the QRHEV2V-DRL approach portrayed a superior accuracy value of 99.62% over existing models under the ToN-IoT dataset.</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

Quantum-resistant hybrid encryption framework for secure and intelligent Vehicle-to-Vehicle communication using deep representation learning models

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月7日

查看全文
  • Tawfiq Hasanin,
  • Zahyah H. Alharbi,
  • Majdy M. Eltahir,
  • Radwa Marzouk,
  • Khalid F. Alsirhani,
  • Safa Alsafari,
  • Turke Althobaiti,
  • Sultan Almutairi

摘要

Vehicular communications are applied in different security and business applications in the digital world for the user’s advantage. During this, vehicle-to-vehicle (V2V) communication allows clients to exchange necessary information with neighbouring vehicles directly. Generally, vehicles move quickly on the highway instead of the crossing road environment, and therefore, a strong method is needed to communicate securely and efficiently. Vehicular Ad Hoc Networks (VANETs) have several distinct features, comprising fast movable nodes, scattered networks, often changing topologies and self-organisation. A malicious attack changes the packet information in the fabrication attack, causing jamming and higher delays in the vehicular network. In this study, a Quantum-Resistant Hybrid Encryption for Secure Vehicle-to-Vehicle Communication Using Deep Representation Learning (QRHEV2V-DRL) method is proposed. The primary aim of the QRHEV2V-DRL method is to demonstrate significant potential in the development of V2V communication networks to ensure long-term data security in the face of evolving cyber threats. In this V2V communication stage, the QRHEV2V-DRL method performs cluster formation using the Kernel Fuzzy C-Means (KFCM) model to partition the communication network into distinct clusters based on similarities in in-vehicle communication patterns. In this attack detection stage, the QRHEV2V-DRL approach utilises min–max normalisation to scale the features within a particular range. Afterwards, the stacked sparse autoencoder (SSAE) model is used for attack classification. In the encryption stage, the quantum-resistant hybrid encryption (QRHE) model is employed to transmit the data in the CH securely to the cloud. The comparison analysis of the QRHEV2V-DRL approach portrayed a superior accuracy value of 99.62% over existing models under the ToN-IoT dataset.