Behavior Mining for Aerial Target Based on Multi-Variant Density-Based Spatial Clustering of Applications with Noise Model

The rapidly increased aircraft types and missions raise difficulties and challenges for aerial target monitoring. In addition, a single sensor usually can only obtain a certain number of attributes of aerial targets within the limited detection range and capability, which makes it difficult to identify the targets. In this paper, a multi-variant density-based spatial clustering of applications with noise model is proposed to mine the behaviors of cooperative and non-cooperative targets under a certain number of historical flight trajectories. Compared with those models that only apply Euclidean distance as the similarity assessment metric, our model can achieve a more accurate and reasonable result on trajectory analysis. In order to verify the applicability, we make a comparison between the mined patterns and the Standard Instrument Departure procedure of certain airport. It demonstrates that our model can precisely identify the patterns of aerial targets.