Wormhole Detection Scheme With Adaptive Deep Neural Network And Hybrid Multi-Objective Mitigation For Internet Of Things
Year: 2024
Type: article
Abstract: The Internet of Things (IoT) is considered a significant innovation by researchers and industry leaders, offering significant advancements in daily life through the use of smart objects that communicate with one another over the internet. Moreover, because IoT devices are resource-constrained and low-cost, implementing an effective security model has emerged as a major technical challenge in recent years. Among the security threats to wireless networks, the wormhole attack is a particularly important and vulnerable issue, as it can disrupt various routing protocols through its activation in various modes. In the transmission of data from a source node to a destination node in network, compromised nodes, also known as wormhole nodes or attackers, may leak or alter the routing information of a specific path. These paths, which contain malicious nodes, are referred to as wormhole paths. To enhance security in IoT, it is crucial to identify and secure such paths. This research presents a novel model aimed at detecting and preventing wormhole attacks to ensure the secure transmission of data in IoT networks. The proposed model comprises two main phases: (a) the detection phase which utilizes an adaptive Deep Neural Network (DNN) based on the Crow Search with Bird Swarm Optimization (CS-BSO) algorithm to detect wormhole attacks across k-different paths, and (b) the prevention phase which utilizes the CS-BSO algorithm in combination with the Bird Swarm Algorithm (BSA) and Crow Search Algorithm (CSA) to determine the optimal path for data transmission. This optimal path selection is based on a multi-objective function that considers network parameters such as delay or latency, energy, packet loss ratio, distance, and throughput, effectively protecting the network from wormhole attacks.
Authors: R. Chandra Sekhar Reddy, A G Mallikarjuna, Mamilla Rathaiah, B Srinivasa Reddy, Emmadi V Raghava, T. Aditya Sai Srinivas, Ashok Sarabu, Gunisetty Ramasekhar, Sangapatnam Suneetha
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4405887653