Agent-Based Wormhole Attack Detection and Prevention Algorithm in the Cloud Network Using MapReduce Technique

Abstract

In day-to-day life, various cloud-based services are being used for disparate purposes because of its perpetuity and diverse dexterity. The cloud computing system has some affinity for distributed systems as both use the various features of networking. While using the cloud-based services, a colossal amount of data comes into picture, which upsurges the data traffic in the network. A cloud environment for this huge amount of data, it is challenging to secure it from various kinds of attacks. So an extensible and decisive threat monitoring, detection and prevention system is required for providing a highly securable infrastructure. The research have been done so far focused only monitoring the network for malicious activities, but not detecting and preventing any particular attack. This paper concentrates on a special attack called a wormhole attack on the cloud computing network. The paper proposes a scheme which detects and prevents the wormhole attack using agent nodes in the network. Multiple agent nodes having IDS capability are used to monitor the behavior of other nodes by analyzing the network traffic. Analysis of traffic is done from the following parameters: hop count, time delay, data packet’s path. In cloud network, since the huge amount of traffic is captured by IDS, the work incorporates MapReduce technique to make the approach effectively. The Map procedure concurrently processes the data using the key value pair <k, v>, in our case key values are the parameters used for traffic analysis. The reduce procedure works on the basis of threshold defined for the parameters produces the list of nodes, which are creating wormhole attack in the network. The cornerstone of the proposed approach is, it does not impose any burden on the server, effective for all cloud models. For prevention of the wormhole attack, ID’s of malevolent nodes is broadcasted to other nodes that the given nodes are detected as wormhole nodes.