Improvised MANET Selfish Node Detection using Artificial Immune System based Decision Tree

Abstract

Mobile Ad hoc networks (MANET) are a grouping of independent mobile devices that can act as both host and router. This type of network does not require a centralized controller, can be created or dissolved dynamically, and does not necessarily include fixed infrastructure. MANETs encounter packet delays and loss due to the presence of nodes with malicious intent. Nodes that deviate from normal behavior are termed as selfish nodes. A selfish node is responsible for packet loss, often to conserve energy, and this harms traffic flows within the MANET. MANET nodes take on the responsibility for the relay and routing of packets being transmitted across the network and it is vital that the participating nodes remain unselfish. However, nodes may become selfish as resources diminish, particularly energy, and it is important to identify selfish nodes so that they can be avoided from the routing path. Here an improved scheme based on an Artificial Immune System incorporating a Decision Tree is used to identify selfish nodes. This strategy improves MANET packet delivery performance.