Network Intruder Detection and Cyber Attack Prediction System

Wireless networks have undergone significant advancements, enabling users to access the internet without the need for physical connections to a router. This flexibility has facilitated the widespread adoption of Internet of Things (IoT) devices, such as smartphones, drones, and cameras, which rely on wireless technologies like Infrared, Bluetooth, IrDA, and IEEE 802.11 for seamless connectivity. These devices often operate in complex configurations, allowing hundreds to thousands of users to connect simultaneously, thereby enhancing performance and driving high-profit margins. However, the proliferation of IoT networks has also introduced significant security challenges. Wireless networks are increasingly vulnerable to dangerous web attacks, which pose a substantial threat to IoT ecosystems. IoT networks are particularly sensitive to such attacks, and even simple Denial of Service (DoS) attacks can disrupt entire networks, rendering devices unable to provide their intended services. This is especially concerning when IoT networks support critical applications or perform essential functions. Despite considerable efforts to enhance IoT security, the success rate in preventing web attacks remains insufficient. One promising approach to improving network security is the implementation of Intrusion Detection Systems (IDS). These systems can leverage data mining techniques to detect and mitigate web attacks effectively. Among the various machine learning algorithms available, the XGBoost algorithm has emerged as a powerful tool for enhancing the accuracy and efficiency of IDS.