No abstract

The security of the network is a significant issue in any distributed system. For that intrusion detection system (IDS), have been proposed for securing the network from malicious activities. This research is proposed to design and develop an anomaly detection model for detecting attacks and unusual activities in IoT networks. The primary objective of this research is to design efficient IDS for IoT network. The intrusion detection plays an essential role in detecting different attacks on IoT and enhances the performance of the IoT. In this research, anomaly detection in IoT networks using glowworm swarm optimization (GSO) algorithm with principal component analysis (PCA) is proposed. However, the proposed model is metaheuristic algorithm-based anomaly detection model to identify attacks by using the NSL-KDD dataset. The GSO algorithm based on PCA is implemented to perform the anomaly detection. For feature extraction, the PCA is used, and for classification, the GSO algorithm is used. For performance analysis, various parameters like accuracy, precision, recall, detection rate and FAR are evaluated. For normal class the proposed model achieved 94.14% accuracy, for DoS 95.52%, for R2L 93.15%, for probe 93.50% and for U2R 88.62% accuracy. Overall the detection rate was 94.08% and FAR was 3.41%.

No abstract

All articles must contain an abstract. Peer to Peer (P2P) network is useful for developing applications and downloading files from the internet. In the existing P2P networks, communication is initiated and supported by peers. Security issues constitute the main challenge in the design of reliable communication techniques for P2P networks. In the past, Rumor Riding Protocol was used in distributed systems for performing coordination and communication. However, the existing Rumor Riding Protocol is used flooding technique to forward the response to the initiator it may leads high network traffic to avid this I have Enhanced Multi path forwarding technique to Rumor Riding Protocol and also enhancement with security features to make them suitable for downloading applications in P2P networks. In this method, it enhances the Rumor Riding Protocol with security features by including a Cryptographic Puzzle, a Challenge Question scheme using trust modeling and an intelligent beta reputation model for secured communication in P2P networks. For this purpose, five new techniques are proposed in this research work, all of which aim at enhancing the security of communication using Rumor Riding Protocol in the P2P networks.

Mobile ad hoc network (MANET) is a special type of wireless network in which a collection of wireless mobile devices (called also nodes) dynamically forming a temporary network without the need of any preexisting network infrastructure or centralized administration. Currently, Mobile ad hoc networks (MANETs) play a significant role in university campus, advertisement, emergency response, disaster recovery, military use in battle fields, disaster management scenarios, in sensor network, and so on. However, wireless network devices, especially in ad hoc networks, are typically battery-powered. Thus, energy efficiency is a critical issue for battery-powered mobile devices in ad hoc networks. This is due to the fact that failure of node or link allows re-routing and establishing a new path from source to destination which creates extra energy consumption of nodes and sparse network connectivity, leading to a more likelihood occurrences of network partition. Routing based on energy related parameters is one of the important solutions to extend the lifetime of the node and reduce energy consumption of the network. In this paper detail literature survey on existing energy efficient routing method are studied and compared for their performance under different condition. The result has shown that both the broadcast schemes and energy aware metrics have great potential in overcoming the broadcast storm problem associated with flooding. However, the performances of these approaches rely on either the appropriate selection of the broadcast decision parameter or an energy efficient path. In the earlier proposed broadcast methods, the forwarding probability is selected based on fixed probability or number of neighbors regardless of nodes battery capacity whereas in energy aware schemes energy inefficient node could be part of an established path. Therefore, in an attempt to remedy the paucity of research and to address the gaps identified in this area, a study examining the impact of residual energy on forwarding probability and how to exclude small amount of residual battery level during route establishment are critically important.