Abstract
Intrusion Detection Systems (IDS) and Intrusion Prevention Systems (IPS) are critical security solutions designed to monitor and manage network activities in order to detect and respond to attacks or abnormal behaviors. An IDS primarily monitors network traffic and generates alerts upon detecting suspicious activities, whereas an IPS can proactively respond to threats by automatically blocking packets, terminating connections, disabling accounts, or isolating devices from the network. However, traditional IDS/IPS systems are limited in their ability to detect complex anomalous patterns in network traffic and often respond ineffectively in dynamic and heterogeneous network environments. This research proposes an intelligent IDS/IPS framework for LAN environments, integrating a Deep Learning‑based Autoencoder to perform anomaly detection on network traffic features. As an unsupervised learning approach, it enables the identification of anomalies without requiring pre‑labeled attack data, while an automated alert mechanism and real‑time response capability enhance the effectiveness of threat detection, monitoring, and prevention in a more proactive and efficient manner.