Abstract
The detection of events in monitored areas is one of the most common tasks employed in military and security sensor applications. Magnetic sensors represent a part of a multisensor system and they are mainly dedicated for a detection of variety classes of vehicles. The vehicle movement has an impact on the Earth's magnetic field in monitored area; therefore this magnetic perturbation can be measured by appropriate magnetic sensor systems. The magnetic signals as a feature describing the detected vehicle consist not only of the signal of interest, but also the additional noise and electromagnetic interferences. The successful detection of vehicles or other events depends on the ability to suppress undesired interferences and reduce the noise. Some sensor applications require not only detection of vehicles base on their magnetic features, but also classifying them into specific classes. The feature generation, as one of classification process steps, requires the reduction of redundant data and at the same time optimization of algorithms from the computation complexity point of view. The paper presents results of experiments with designed magnetic sensor system based on the fluxgate magnetometer Mag566 and deals with magnetic signal processing algorithms for the noise and interference reduction. In conclusion, results of the feature generation process for vehicles detection are presented.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have