High-resolution through-wall radar imaging method based on group structure and correlation learning

Abstract

The invention discloses a high-resolution through-the-wall radar imaging method based on a group structure and correlation learning, which relates to the technical field of radar imaging. The method comprises the following steps: a uniform linear array of N elements is assumed, and direct echo signals of the nth element are simulated; N-element M-frequency point echo signals are vectorized at thesame time; N-element multi-path echo signals are simulated; the compressed observation data are acquired; a hierarchical Bayesian generation model is constructed; variables gamma, x and c are optimized; gamma, x and c are initialized; the objective function reduction upper bound as described in the specifications after a non-zero element of gamma is changed to a zero element is calculated; the objective function reduction upper bound as described in the specifications after a zero element of gamma is changed to a non-zero element is calculated, and gamma is converged to the optimal value, wherein the parameter c is the learnt intra-group element correlation. The method disclosed in the invention makes full use of the internal group sparse structure of the signals and the correlation between intra-group elements, the sparse reconstruction mean square errors of the signals can be effectively improved, the number of observations needed to reconstruct the original signals is reduced, and the anti-noise ability is stronger.

Description

technical field [0001] The invention belongs to the field of radar imaging, in particular to a high-resolution radar imaging method based on group structure and correlation learning. Background technique [0002] In recent years, with the increasing demands of public security departments, fire disaster rescue personnel and rapid response defense forces to obtain information on areas behind walls and inside rooms in urban areas, through-wall radar imaging, as a new generation of perspective imaging technology, has received a lot of attention from the scientific research community and A lot of attention and research in the industry. Through-the-wall radar imaging adopts ultra-wideband microwave radar technology to realize the penetration detection of non-electromagnetic transparent media such as walls, trees, bushes, and smoke through electromagnetic waves, and complete the panoramic imaging display of the target area behind the barrier, including the detection target image, a...

AI Technical Summary

Problems solved by technology

However, these methods are based on the ray model and do not consider the problem of multipath propagation, and such group sparse reconstruction methods are sensitive to the correlation of the perceptual dictionary matrix, and the high resolution determines the high correlation of the perceptual dictionary matrix. This will affect the refactoring performance of methods of this class

Images