Active Learning Artificial Neural Networks Ensemble for HF Ground Wave Radar Sea Clutter Predicting

Based on chaotic characteristic of high frequency ground-wave radar (HFGWR) sea clutter, a new adaptive artificial neural networks ensemble method for sea clutter predicting is presented in this paper. In phase space reconstructed, when one sea clutter sample is to be predicted, some artificial neural networks are choosed adaptively by evaluating their performance and error correlation in neighborhood of the sample, and outputs of these artificial neural networks are combining dynamically as the result of prediction for the sample. The adaptive artificial neural networks ensemble method is designed to improve precision of sea clutter predicting, and server sea clutter modeling in HF ground wave radar objects detecting. In order to improve the adaptive ability of the ensemble method and reduce computational complexity, the corresponding active learning algorithm is designed. Result of testing the sea clutter predicting method on practical echo data of HFGWR for objects detecting shows precision of sea clutter predicting and generalization ability can be impoved effectively by the adaptive artificial neural networks ensemble method.