Machine learning approaches on hardware based intrusion detection system and implementations of 802.11S attacks on an 802.11AC based wireless testbed environment

Application areas of wireless LAN technologies follow a rapid expansion trend due to the ever increasing bandwidth they offer at the physical layer. This trend inevitably entails higher costs due to potential cyber security vulnerabilities. In the first part of this thesis, end-to-end and modular Wireless Intrusion Detection System are implemented with the machine learning approaches on hardware in 802.11 Basic Service Set architecture. In the second part, Fake Mesh Authentication, Path Diversion and Blackhole attacks which are defined in 802.11s Wireless Mesh Networks are implemented using software architecture specifically created for these attacks.