Malicious Nodes Identification and Classification of Nodes and Detection of UDP Flood Attack with ICMP using OLSR Routing Protocol in MANET
Sweta Kriplani, Rupam Kesharwani · Zenodo (CERN European Organization for Nuclear Research) · 2016
Mobile ad hoc networks also known as MANETs have been used extensively for the seamless provisioning of information exchange, where the deployment of infrastructure is difficult, if not impossible. Such cases include remote rural areas with stringent topographical profiles, disaster-recovery terrains, battlefields and popular event sites (i.e. sports stadiums, exhibition venues). This new approach of networking brings a great flexibility and affordability to the world of wireless communications by introducing pervasive computing, document sharing, and smart sensors. However, since this kind of network uses the wireless medium for communication, the wireless ad-hoc network faces several security risks at different layers. A particularly severe security attack that affects the ad hoc network routing protocols at the network layer, is known as the Black hole attack. A malicious node advertises itself as having the freshest or shortest path to destination. Once the malicious node attracts the traffic toward itself, the attacker can misuse or discard the traffic and as a result data through the malicious node is lost. To properly protect these systems with limited resources, the security practitioners need to understand the possible security threats and their impacts on MANETs and have a framework to ensure that the protections implemented to mitigate the vulnerabilities in the systems are the most efficient ones possible. In this Research the effects of malicious nodes on MANETs proactive routing protocol, Optimized Link State Routing (OLSR) were studied using NS-3 and their vulnerabilities compared. A framework for a methodical security analysis and recommendation of efficient protection schemes was also developed as well Support Vector Machine (SVM) for classification of categories of node as malicious and normal nodes and Apply filtering rule on network traffic for identification of UDP FLOOD and ICMP attack using Wireshark.