ارمان سميعي

With the development of Internet of Things (IoT) technology and the widespread use of network connected devices, cybersecurity has become one of the fundamental challenges in protecting users’ data and personal information. Due to hardware and software limitations, such as weak processors and limited storage space, these devices are prone to complex cyberattacks. On the other hand, IoT networks, having numerous points of vulnerability and a large number of connected devices, have become an attractive target for attackers. One of the key methods to counter these threats is the use of intrusion detection systems, which help identify and prevent network attacks. However, the limited processing resources of these devices pose challenges to deploying high performance intrusion detection systems. In this research, a novel solution is proposed by combining the Horse Herd optimization algorithm and the nearest neighbor algorithm to improve intrusion detection in IoT networks. By selecting optimal features, the Horse Herd optimization algorithm enhances detection accuracy and reduces the complexity of the intrusion detection system. This model was eva‎luated on two datasets NSL KDD and UNSW NB15, and the results indicate detection accuracies of 98.86 percent and 98.88 percent respectively, while also showing a significant reduction in the false positive rate compared to previous studies. This study demonstrates that the combination of optimization algorithms and classification methods offers an effective approach to address security challenges in IoT networks. Optimizing computational resources, achieving higher speed in attack detection, and enabling the implementation of such systems on devices with limited resources are among the key advantages of this method. These findings highlight the high capability of this approach in enhancing the security of IoT devices.

اينترنت‌اشيا , سيستم‌هاي تشخيص نفوذ , الگوريتم بهينه‌سازي گله اسب

Internet of Things , Horse Herd Optimization Algorithm , Intrusion Detection Systems

Arman Samiee

Hadi Shahriar Shah Hosseini