Elevating Security Measures in Ad Hoc Networks: An Intrusion Detection Approach with Deep Learning

Ad hoc networks are critical in contexts like disaster recovery and military operations, where standard infrastructure is either unavailable or unfeasible. Despite their versatility, these networks are naturally vulnerable to routing-based attacks such as the black hole attack, which involves rogue nodes intercepting and discarding data packets. This paper tackles this problem by presenting a deep learning-based intrusion detection system (IDS) that uses the VGG16 architecture. This study employs computer modeling to simulate such attacks and proposes a novel intrusion detection system based on machine learning algorithms, particularly utilizing the VGG architecture. The system aims to categorize network packets as safe or dangerous, enabling the identification of intrusions. Through experimentation, it is demonstrated that this method shows promise across various classifiers and can adapt to evolving attack strategies. The need for robust detection mechanisms persists amidst continuous changes in attack methodologies. The simulation and execution of the proposed system are implemented using MATLAB software.