DSN-based high-resolution two-dimensional ISAR imaging method

Abstract

The invention discloses a DSN-based high-resolution two-dimensional ISAR imaging method. The method solves the problems that the 2D-FISTA algorithm has openness of optimal regularization term coefficient selection, a CV-DNN network lacks theoretical support and needs high space and time complexity, and DSN cannot solve the problems of main lobe width, high sidelobe and the like of an RD image. Themethod comprises the following implementation steps: obtaining wave number domain echoes in a two-dimensional matrix form and wave number domain echoes in a one-dimensional vector form of ISAR two-dimensional scattering point distribution; solving a one-dimensional vector form and a two-dimensional matrix form of two-dimensional scattering point distribution; building a DSN network; setting a loss function; performing DSN network training; and completing the ISAR high-resolution two-dimensional imaging based on the DSN. Based on a sparse signal reconstruction theory, an SALSA algorithm is constructed into a deep network, high-resolution two-dimensional ISAR imaging is realized, an ISAR image with good focusing and a clean background is obtained, and the method can be used for efficientlyperforming high-resolution two-dimensional ISAR imaging in batches in complex electromagnetic environments with target echo defects, noise and the like.

Description

technical field [0001] The invention belongs to the technical field of radar, and further relates to radar signal processing, in particular to a high-resolution two-dimensional ISAR imaging method based on DeepSALSA-Net (DSN), which is used for high-resolution two-dimensional imaging of ISAR. Background technique [0002] Inverse synthetic aperture radar high-resolution imaging plays an important role in space situational awareness and air target surveillance. In an ideal observation environment, for a stable target, ISAR can obtain its high signal-to-noise ratio echo, and then through the existing classic algorithms, such as Range-Doppler algorithm (Range-Doppler, RD), polar format algorithm (Polar Formatting Algorithm , PFA) etc. to obtain well-focused two-dimensional high-resolution ISAR images. However, when the target radar cross-sectional area is small or the observation distance is long, the signal-to-noise ratio of the echo is limited by the transmission power. At ...

AI Technical Summary

Problems solved by technology

However, the disadvantage of this method is that the selection of the optimal regularization coefficient is still an open issue, and if the selection is inappropriate, the optimal imaging result cannot be obtained.

However, the disadvantage of this method is that its network design is highly subjective and there is no uniform standard

However, the disadvantage of this method is that it cannot effectively solve the inherent main lobe width and high side lobes of the RD image.

In addition, the network needs to use the measured data of the same kind of target for training, and it is difficult to meet the network training requirements when the target measured samples are limited by the observation conditions.

[0006] In the prior art, the 2D-FISTA algorithm has an open problem of selecting the optimal regularization term coefficient, and the network of CV-DNN lacks theoretical support and requires high Due to the space and time complexity, DSN cannot effectively solve the inherent main lobe width and high side lobes of RD images.

Images