Search

Sea clutter is a kind of ubiquitous interference in sea-detecting radars, which will definitely influence target detection. An accurate sea clutter prediction method is supposed to be beneficial while existing prediction methods are based on the one-step-ahead prediction. In this paper, a sea clutter prediction network (SCPNet) is proposed to achieve the k-step-ahead prediction based on the characteristics of sea clutter. The SCPNet takes a sequence-to-sequence (Seq2Seq) structure as the backbone, and a simple self-attention module is employed to enhance the ability of adaptive feature selections. The SCPNet takes the normalized amplitudes of sea clutter as inputs and is capable of predicting an output sequence with a length of k; the phase space reconstruction theory is also used to find the optimized input length of the sea clutter sequence. Results with the sea-detecting radar data-sharing program (SDRDSP) database show the mean square error of the proposed method is 1.48 × 10−5 and 8.76 × 10−3 in the one-step-ahead prediction and the eight-step-ahead prediction, respectively. Compared with four existing methods, the proposed method achieves the best prediction performance. [ABSTRACT FROM AUTHOR]

Since solving the path of radio wave propagation by an analytic method in the process of coordinate registration (CR) for skywave over‐the‐horizon radar (OTHR) is difficult, a reference source‐aided CR method that uses ensemble learning is proposed. First, the OTHR wave propagation principle is briefly described, and the ionospheric channel characteristics implied by reference sources and target information are explained, which provides a theoretical basis for modelling. Then, multiple machine learning models, such as random forest, support vector machine and elastic network, are used to build a mapping network between known information and the actual ground distance of the targets. Moreover, the prediction ability and correlation of each machine learning model are evaluated. Finally, with stacking ensemble learning, machine learning models with better performance to solve CR problems are effectively combined as base learners. The models develop the advantages of each base learner to achieve a perception of the ionospheric environment and accurate OTHR CR. The simulation results show that the proposed method can mine the ionospheric channel propagation characteristics and effectively reduce the CR error of the targets of interest. When the information provided by 4 reference sources 200 km away from the target is used, the average error of target positioning is 13 km. [ABSTRACT FROM AUTHOR]