Search

The structure design and switching regulation of pulse groups in multi‐function radars (MFRs) are closely related to the work mode. The sequential extraction and recognition of MFR pulse group structure is a fundamental task to analysing and interpreting the work modes and behaviour intentions of an MFR. In this study, the temporal structure of MFR pulse group is expressed hierarchically, which is intensively modelled based on regular grammar. Besides, a corresponding hierarchical automaton is established to sequentially extract and recognise the MFR pulse group structure in pulse train. The hierarchical automaton used for pulse group recognition has a two‐layer structure. The bottom layer of the hierarchical automaton realises the sequential input of pulses and recognition of pulse subgroups, and the sequential input of pulse subgroups and recognition of pulse groups are realised at the top layer. The simulation results demonstrate that the proposed method performs satisfyingly in recognising the pulse group structure and is robust to not only pulse noises but also to the emitter number and pulse group scale. [ABSTRACT FROM AUTHOR]

In this letter, a deep learning framework for direction of arrival (DOA) estimation is developed. We first show that the columns of the array covariance matrix can be formulated as under-sampled noisy linear measurements of the spatial spectrum. Then, a deep convolution network (DCN) that learns the inverse transformation from large training dataset is introduced. In contrast to traditional sparsity-inducing methods with computationally complex iterations, the proposed DCN-based framework could efficiently obtain DOA estimates in near real time. Moreover, the utilization of the sparsity prior improves DOA estimation performance compared to existing deep learning based methods. Simulation results have demonstrated the superiority of the proposed method in both DOA estimation precision and computation efficiency especially when SNR is low. [ABSTRACT FROM AUTHOR]