Conviction Based Packet Promotion Scheme for Efficient Detection of Selfish Nodes in Mobile Ad Hoc Networks

Abstract

Nodes in the Mobile Ad Hoc Network (MANET) moves arbitrarily because of infrastructure less environment. Since there is no fixed infrastructure, the mobility of nodes causes frequent changes in the topology of the network. Due to the lack of maintaining the centralized access control, MANETs are vulnerable to diverse type of loss such as packet loss, entering of selfish nodes and security threats. A loss of packet mainly happens based on presence of malicious nodes or selfish nodes. A node may act selfishly to preserve its scarce resources and do not forward the packets to other nodes but still want to use their services to send and receive their own packets. Several researchers focused on detecting and preventing the entry of selfish nodes in the network. Existing work presented an approach to verify the packet forwarding by the Merkle tree principle approach. But the mobility on packet data in mobile ad hoc networks needs an extension to improve the network lifetime. So in this work, a Conviction Based Packet Promotion Scheme (CBPPS) is introduced for detecting the selfish nodes using the direction-finding level behavior information. It uses conviction values to favor packet promoting by maintaining a conviction counter for each node which increases throughput and decreases the communication overhead. The detected selfish node is punished or rewarded by decreasing or increasing the conviction counter. Simulation experiments are conducted with various simulation scenarios to estimate the performance of the proposed CBPPS scheme for prolonged network lifetime against merkle tree principle [1] in terms of percentage of selfish nodes detected, time of detection, average packet delivery ratio. With CBPPS scheme, the selfish node detection effectiveness is high compared to existing merkle tree principle approach
Copyright © 2014 Praise Worthy Prize - All rights reserved.