Enhanced intrusion detection with fiber optic sensors in rainy weather conditions: a comparative study of phase parameter matching and cross rebuilding algorithms for higher end communication

Fiber optic intrusion detection is widely regarded as one of the most effective perimeter defense strategies against intruders or terrorists. Currently, most infiltration detection systems gather data through optical cables and then track intrusions using classification techniques. In this study, we present two intrusion detection algorithms that utilize system recognition to tackle the challenge of unstable fiber optic infiltration features, especially during rainy weather conditions. The primary method involves the computation of the system's disturbance covariance at any given position. This process serves as a foundational step in identifying intrusions within the system. Utilizing the Frobenius norm, we quantify the disparity between the disturbance covariance matrices obtained from different sites. By assessing these differences, we can effectively discern the presence of an intruder and pinpoint their location within the system. The second method employs the Kalman filter to rebuild the output of each point the non-infiltration point state parameters, and in order to detect intrusion, the cross-reconstruction error is determined. Data from experimental tests is used to evaluate the two methods. The experimental outcomes indicate that the proposed methods are both viable. When comparing, the first approach is more solid, while the second is faster when computing.