Evaluation of Machine Learning Algorithms for Intrusion Detection System in WSN

Technology has revolutionized into connecting “things” together with the rebirth of the global network called Internet of Things (IoT). This is achieved through Wireless Sensor Network (WSN) which introduces new security challenges for Information Technology (IT) scientists and researchers. This paper addresses the security issues in WSN by establishing potential automated solutions for identifying associated risks. It also evaluates the effectiveness of various machine learning algorithms on two types of datasets, mainly, KDD99 and WSN datasets. The aim is to analyze and protect WSN networks in combination with Firewalls, Deep Packet Inspection (DPI), and Intrusion Prevention Systems (IPS) all specialized for the overall protection of WSN networks. Multiple testing options were investigated such as cross validation and percentage split. Based on the finding, the most accurate algorithm and the least time processing were suggested for both datasets.

@article{Alsahli2021,
  title     = {Evaluation of Machine Learning Algorithms for Intrusion Detection System in WSN},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {12},
  number    = {5},
  year      = {2021},
  publisher = {The Science and Information Organization},
  author    = {Mohammed S. Alsahli and Marwah M. Almasri and Mousa Al-Akhras and Abdulaziz I. Al-Issa and Mohammed Alawairdhi},
  doi       = {10.14569/IJACSA.2021.0120574},
  url       = {https://doi.org/10.14569/IJACSA.2021.0120574}
}