OntoIDF: a smart knowledge-based framework for intrusion detection in autonomous cloud computing environment

OntoIDF: a smart knowledge-based framework for intrusion detection in autonomous cloud computing environment

연구 분야: Infrastructure

학회: International Journal of Information Technology

Wireless Sensor Networks (WSN’s), the part of the cloud computing environment, are equipped with several applications and are susceptible to attacks by intruders. Existing intrusion detection systems (IDS) are able to identify simple attacks instead of complex attacks. Complex attacks comprise multiple forms of attacks that can only be identified through reasoning and maintaining relationships between sensor nodes and patrol nodes in the network. Therefore, the given paper proposes a smart knowledge-based framework to detect intrusions in the given network with the use of ontology. The proposed framework begins with the capturing of live network packets, followed by the construction of ‘NETWORK TRAFFIC Ontology’ in order to represent the relationship between the captured packets and attributes associated with several nodes. The designed ontology serves as a knowledge base comprising a set of classes, properties, and instances in the context of packets captured from autonomous cloud computing environment. After this, the features from the designed ontology are extracted using semantic web rule language (SWRL) and Pellet reasoner. It is used to identify intrusion in the form of malicious packets in the network. Lastly, the performance of the proposed framework is validated against recent studies based on evaluation metrics such as attack estimation rate (AER), accuracy (%), and false alarm rate (FAR). The results show that the proposed framework outperforms the recent studies in terms of performance metrics.