Mean Donkey and Smuggler Optimization (MDSO) Based Cluster Head Selection and Recurrent Neural Network Clone Node Detection (RNNCND) for Wireless Sensor Network (WSN)

Wireless sensor network (WSN) have been set up in open, unrestricted spaces where intruders can seize sensors and replicate stolen nodes. Clone nodes can conduct a variety of cyberattacks since they are thought of as legitimate nodes. Both fixed and mobile sensor networks include a wide variety of clone detection algorithms. Cluster-based Clone Attack Detection (CBCD), a clone attack detection mechanism, is used in this research to find accessible clones in mobile WSNs. Sensor networks are organised into clusters according by this protocol. A master node and an arbitrary number of sensor nodes are present in every cluster. Mean Donkey selected Cluster Head, and Smuggler Optimisation (MDSO) was added to enhance network performance. Then, to find the clone node that was present in the WSN, the Recurrent Neural Network Clone Node Detection (RNNCND) technique was created. To prevent clone assaults, this entails conducting efficient clone detection. With affordable identity verification, find duplication both locally and globally. Network Simulator 2 (NS2) is extensively used to produce the suggested technique. Finally, performance analysis is carried out, and both new and old methodologies are examined to show the success of the system. The suggested strategy, according to simulation findings, improves packet delivery rate (PDR) and decreases packet loss throughout. Delay and energy use.