Multi-class DOS attacks classification in C4I systems

Abstract

C4I systems faced serious concerns on the security of infrastructures and the integrity of sensitive data. Fast growth of C4I systems in different domains has attracted the researchers to find ways to secure such systems from critical attacks. DOS attacks falls in the category of critical attacks that compromises the availability of the resources and detection of these attacks is also a challenging task. To overcome this problem in the C4I environment a system is proposed. Further, multi-class problem of attack detection is another issue in the C4I systems. Therefore, this work focus on DOS attacks detection and classification in the C4I environment. Multilayer Perception (MLP) is used for classification purpose due to its proven ability in classification. This research work uses the Knowledge Discovery and Data mining (KDD) cup dataset, which is considered benchmark for evaluating security detection mechanism. The performance of this approach was analyzed. The results show that proposed method provides an optimal intrusion detection mechanism in C4I systems.