Prevention of Black Hole Attacks in Virtualized Cloud Network Using Trust-Aware Energy Efficient AODV Routing With Firefly Based AI Technique

Abstract

Cloud networks are very widespread and unreliable because of the amount of VMs and presented nodes in their Virtual Cloud Network. Nodes might connect and revoke networks at any time. Resilience is a advantage of cloud computing, but it has many safety issues in routing and transmitting information between messages. VCN research is very similar to the portable ad-hoc network (MANET), which depends on the collaboration of all involved nodes to provide fundamental activities. Many safety assaults and risks exploit the safety of information transmission due to the decentralized environment in VCN and MANET. Malicious nodes can interfere and use information during wireless communications. Numbers of methods are there that has a diverse effect on such attacks for malicious nodes. Varied attacks are susceptible to security, but Black hole assault is one of the most common effective assaults, as fraudulent nodes dump all incoming emails reducing network performance and reliability. A black hole node is designed to lampoon every node in the network that conveys with some other node by saying it always has the easiest route to the target node. In this manuscript, a secure routing discovery method has been presented using Ad hoc on demand distance vector (AODV) routing protocol. For the detection of attack in the cloud, the concept of Artificial Intelligence (AI) has been used. Therefore, in this research, Artificial Neural Network (ANN) and Support Vector Machine (SVM)is adapted to determine Packet Delivery Ratio (PDR), Delay and Throughput measures. The comparative examination has been conducted to depict the proposed FNN-AODV effectiveness. There is an enhancement of 61.01% in FNN-AODV and 5.08% enhancement in Throughput in proposed FNN-AODV than R-AODV, 6.26% enhancement in PDR for FNN-AODV than R-AODV and 10.8% is the decrement in delay in FNN-AODV than of R-AODV.