A comprehensive review of using optical fibre interferometry for intrusion detection with artificial intelligence techniques

Security remains a critical concern in today's world, especially for protecting high-value assets and vital infrastructure such as refineries, petrochemical plants, government facilities, and military installations. Traditional security measures often fall short against increasingly sophisticated threats. To meet these challenges, perimeter intrusion detection systems (PIDS) have become indispensable. Optical fiber interferometry (OFI), an advanced sensing technology, provides key advantages for PIDS, including high sensitivity, real time monitoring, immunity to electromagnetic interference, and long-range coverage. This research explores the integration of OFI with machine learning and deep learning techniques, enhancing intrusion detection and classification capabilities. Machine learning allows systems to process vast amounts of sensor data, recognize patterns, and accurately classify threats in real time. Deep learning further optimizes this by simulating neural networks to understand complex data relationships, reduce false alarms, and improve adaptive learning. The fusion of these technologies marks a significant leap forward in security, enabling intelligent, responsive, and highly accurate intrusion detection solutions.