Wireless sensor networks (WSN) are networks that are self-configuring and run without any central coordinator. Security and energy consumption are a big problem for wireless sensor networks because of these network properties. The malicious nodes that join the network cause security attacks.In terms of active and passive attacks, it is possible to distinguish attacks. In the active form of attack, the sinkhole assault. The malicious nodes in the sinkhole attack spoof the base station's identity and behave like the base station. Instead of base stations, the sensor nodes start transmitting data to malicious nodes. Cluster heads are chosen on the basis of energy, distance, to minimize energy consumption of the network LEACH protocol, which divides the entire network into clusters and clusters. Therefore, the total life of the network is decreased and the amount of energy consumed is increased when a sinkhole attack occurs.In this research paper, the latest method is designed to identify and isolate malicious nodes from the network. This algorithm works by using recognition to quickly classify attacks. The proposed algorithm is implemented in NS2 and those parameters are evaluated in terms of performance. As opposed to current techniques, our proposed technique performs well in terms of all parameters.

Keywords : Active, Attack, Sinkhole, Malicious, LEACH, WSN