JavaScript is disabled for your browser. Some features of this site may not work without it.
| dc.contributor.author | Esubalew, Mulat | |
| dc.date.accessioned | 2021-10-19T11:34:48Z | |
| dc.date.available | 2021-10-19T11:34:48Z | |
| dc.date.issued | 2021-10-19 | |
| dc.identifier.uri | http://ir.bdu.edu.et/handle/123456789/12802 | |
| dc.description.abstract | Internet of Things (IoT) is an extension of the Internet. It extends the human to human interconnection and intercommunication of the Internet by including things, shifting the current Internet scenario to bring anytime, anywhere, and anything communication. A discipline in networking evolving in parallel with IoT is Software Defined Networking (SDN). It is a significant technology and can solve the different problems existing in the traditional network systems. It provides a new home to address the different challenges existing in different network based systems including IoT. Yet, the issue of security of SDIoT environments is still a headache for security researchers since the vulnerablity space of IoT systems are intensified by the incorporation of SDN having its own security limitations. One important security challenge prevailing in such systems is a guarantee of service availability. The ever increasing denial of service (DoS) attacks are responsible to such service denials. A centralized signature based intrusion detection systems (IDS) is proposed and developed in this work. Three machine learning (ML) algorithms, decision tree (DTree), Random Forest (RF) and Support Vector Machine (SVM) are used in the experiments. A very popular and recent benchmark dataset, CICIDS2017, has been used for training and validating the ML models. An accuracy result of 99.967% has been achieved by using only eleven features on the Wednesday’s release of the dataset. This result is higher than the achieved accuracy results of related works considering the original CICIDS2017 dataset. A maximum cross validated accuracy result of 99.728% has been achieved on the same release of the dataset. These developed models meet the basic requirement of a supervised IDS systems developed for smart enviornments and can effectively be used in different IoT service scenarios. | en_US |
| dc.language.iso | en_US | en_US |
| dc.subject | ELECTRICAL AND COMPUTER ENGINEERING | en_US |
| dc.title | MACHINE LEARNING BASED EFFECTIVE INTRUSION DETECTION FRAMEWORK FOR SDN ORCHESTRATED INTERNET OF THINGS | en_US |
| dc.type | Thesis | en_US |
