An Enhanced Heterogeneous Local Directed Acyclic Graph Blockchain With Recalling Enhanced Recurrent Neural Networks for Routing in Secure MANET-IOT Environments in 6G

【Author】 Begum, M. Baritha; Suganthi, B.; Sivagamasundhari, P.; Arunmozhi, S. A.; Suhail, S. J. Muhamed

【Source】INTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS

【影响因子】1.882

【Abstract】Mobile ad hoc networks (MANETs) integrated with the Internet of Things (IoT) form a decentralized communication framework crucial for 6G environments. However, ensuring secure and efficient routing in such networks remains a challenge due to their distributed nature and vulnerability to attacks. This paper introduces a heterogeneous local directed acyclic graph blockchain (HLDAG-BC) combined with recalling enhanced recurrent neural networks (RERNNs) for secure and efficient routing in MANET-IoT environments. The HLDAG-BC offers tamper-proof communication and identity-based conditional privacy-preserving authentication (ICPA) is a lightweight and secure node authentication scheme. Network nodes are grouped using the kernel neutrosophic c-means (KNCM) algorithm. The optimal cluster heads are chosen using the red piranha optimization (RPO) method. RERNN determines the shortest routing path to increase reliability and minimize latency. Furthermore, an HDLNN is used for intrusion detection to achieve robust network security. The HLDAG-BC-RERNN approach proposed shows that the packet delivery ratio improves by 31.35%, throughput by 34.56%, latency by 30.29%, and network lifetime by 28.67% compared to the existing approaches, as shown in the comprehensive evaluations. In conclusion, the proposed framework offers a scalable and secure solution for MANET-IoT networks, making it a viable approach for future 6G applications.

【Keywords】blockchain; deep learning; Internet of Things; intrusion detection; malicious attacks; mobile ad hoc networks; node authentication; privacy-preserving; security

【发表时间】2025 MAR 10

【收录时间】2025-02-23

【文献类型】

【主题类别】

--