Design of Clustering Enabled Intrusion Detection with Blockchain Technology

Abstract

Recent advancements in hardware and networking technologies have resulted in a large growth in the number of Internet of Things (IoT) devices connected to the Internet, which is likely to continue growing in the coming years. Traditional security solutions are insufficiently suited to the IoT context due to the restrictions and diversity of the resources available to objects. Security techniques such as intrusion detection and authentication are considered to be effective. Additionally, the decentralised and distributed nature of Blockchain technology makes it an excellent solution for overcoming the security issue. This paper proposes a chaotic bird swarm algorithm (CBSA)-based clustering technique based on an optimum deep belief network (ODBN) and Blockchain technology for secure authentication in an IoT setting. The CBSA-ODBN technique creates a clustering algorithm that utilises CBSA to pick cluster heads (CHs). The ODBN model is then utilised to identify the network, and the learning rate of the Deep Belief Network (DBN) model is optimally adjusted using the flower pollination technique (FPA). The suggested concept creates a layered security network paradigm for the Internet of Things using blockchain technology. Numerous simulations are run, with the outcomes analysed using a variety of measures, including detection rate, packet delivery ratio, energy usage, end-to-end latency, and processing cost. A careful comparison of the suggested model's performance to recently developed methodologies demonstrated the proposed model's superior performance.