Secure and Energy Efficient Routing in Wireless Sensor Network using Machine Learning

Security and energy efficiency are significant issues in Wireless Sensor Network (WSN). It is a critical task to provide reliability during data transmission in routing. Due to the presence of routing attacks in the network layer, security of the network degrades. A secured and energy-efficient path is necessary to transmit information from the source node to the destination node. It motivates the authors to propose a novel solution that improve the performance of the network This paper is mainly emphasizing an energy-efficient and secure routing mechanism. An optimum cluster head helps to save energy in multi-hop communication. The selection of cluster head depends on the trust factor evaluated through machine learning. The decision tree algorithm is applied to the input variables such as degree of mobility, vulnerability index, residual battery energy and node's distance from cluster centroid. Class acts as a target variable which helps to determine the optimal cluster head. This efficient cluster head determines the shortest path across the network. Security of routing path is enhanced through intrusion detection and removal mechanism. Node mobility mechanism is used for the removal of the malicious node.