Volume 18, No. 6, 2021

Security Analysis Of Malicious Attacks In MANET Through Machine Learning Algorithm

Surabhi Srivastava , Chandra Shekhar Yadav , Pradeep Kumar


There are more nodes in this network that are vulnerable to denial-of-service (DoS) attacks because to the network's more complicated and frantic routing method, making the topology of MANET more volatile than other networks. For example, AODV is more popular than table-driven routing, which relies on flooding to discover the best route. Attackers have taken use of this idea to launch denial-of-service attacks (DoS) similar to floods; the black hole and grey hole attacks are the MANET-branded ones. Network form flexibility and movable node mobility are fundamental aspects of MANETs, which are distinct from other types of networks. Network speed, latency and packet transfer rate are the topic of this essay. A neural network known as a jump field is used to transmit packets. Both end-to-end delays and packet transfer rates and throughput improve. Additionally, there is a section on how to recover from wireless mobile node network congestion. Machine learning applications may prevent packet loss in the future by iterating on the iteration that began. Embedded network applications are discussed in general terms after a look at the research outcomes. There was a lot of emphasis on the in-network processing approaches that were selected and compared to the Hopfield neural network and the back propagation network based on their physical appearance. The number of mobile nodes in a network may be increased. In the next neural network, a new context is introduced. Our test implementation has produced encouraging results so far and we need to describe how neural networks might be employed in a mobile node network.

Pages: 5866-5882

Keywords: MANET, Denial-of-Service Attack, AODV, Neural Networks, Gray hole Attack

Full Text