Application of the HHT method to the airport fiber fence warning

For the airport fiber fence warning system has high false alarm rate, an method to identify invaders accurately is needed urgently at present,. To address the issue, a recognition algorithms using neural network base on the Hilbert-Huang Transform was proposed. Firstly, the common intrusion signals around the airport were decomposed into a set of intrinsic mode function (IMF) using EMD, then the Hilbert transform is applied to each IMF, and the feature is more clearly exhibited by the marginal spectrum. Finally, BP neural network is used to identify the invasion behavior by marginal spectrum. Experiments show that, the accuracy rate based on the method proposed here is more than 95%.