Search

In this paper, a novel interference mitigation approach using an autoencoder in combination with a traditional interference detection filter is introduced. It is shown that by employing the gated convolution, the encoder has the ability to learn the signal pattern from the remaining interference-free signal. The decoder can recover the interference-contaminated signal segments from the bottleneck representation as computed by the encoder. Experimental results show that the proposed method can provide a remarkable improvement in signal-to-interference-plus-noise ratio (SINR) and preserves its robustness on real radar measurements in severely disturbed scenarios that are more complex than the training dataset.

This paper introduces a novel automotive radar interference mitigation approach using a generative adversarial network (GAN). Instead of tackling the mutual interference in the time domain, a generative adversarial network is trained and used to recover the complex signal in the frequency domain, namely on the complex range profile obtained after the fast Fourier transform of fast-time samples (RFFT spectrum). It is shown that by employing the gated convolution and an attention mechanism, the generator network has the ability to learn the amplitude and phase information for missing data from the remaining signal. Experimental results show that the proposed method can provide a remarkable improvement in signal-to-interference-plus-noise ratio (SINR) and preserves its robustness in severely disturbed scenarios that are much more complex than the training dataset.

Infrared measurement is one of important means for pelagic space instrumentation. Abnormal pixels would impair imaging quality of infrared detector and bring bad influence for display enhancing. In this paper, it presents an abnormal pixel detecting method combing statistics and frequency information and an improved display enhancing method tolerating abnormal pixels based on the detecting result. Real image experiment result shows that it could effectively detect abnormal pixels and improve display enhancing performance.

For through-wall radar imaging (TWRI), the building layout determination is an essential task in practice applications. In this paper, two multi-input multi-output (MIMO) radar with 8 transmitting antennas and 2 receiving antennas under two vertical views are deployed to finish detection, and the generative adversarial nets (GAN) is applied to finish building layout determination. The electromagnetic simulations demonstrate that the GAN achieves great performance in building layout determination.

In synthetic aperture through-the-wall radar imaging, wall penetration effect causes target images defocused and displaced from their true positions. To solve this problem, a wall compensation method based on generative adversarial nets (GAN) is proposed in this paper. Specifically, GAN constructs a spatial structure mapping from the original input images to the output images without wall penetration effect. Electromagnetic simulation show that the proposed method has acceptable computational load and great imaging results in both refocusing and correcting the displacement.

For multi-input multi-output (MIMO) through-wall radar imaging (TWRI), multipath ghosts, side/grating lobe artifacts and wall penetration effect degrade the imaging quality of the obscured targets inside an enclosed building, there in hindering target detection. In this paper, an approach based on generative adversarial nets (GAN) is proposed to achieve multipath ghosts, side/grating lobe artifacts and wall penetration effect suppression with regard to MIMO TWRI. Specifically, the whole task is divided into two steps (Firstly, multipath ghosts and wall penetration effect are suppressed but the side/grating lobes are preserved. Secondly, side/grating lobes are eliminated.) Then a GAN network is applied to solve those two steps. Extensive electromagnetic simulations and comparisons demonstrate that the proposed approach achieves better suppression of multipath ghosts, side/grating lobe artifacts, wall penetration effect and other significant superiorities, including priori wall information not being required and robustness for different array deployments and building layouts.

In this paper, we propose an approach that uses generative adversarial nets (GAN) to eliminate multipath ghosts with respect to through-the-wall radar imaging (TWRI). The applied GAN is composed of two adversarial networks, namely generator G and discriminator D. The generator G learns the spatial characteristics of an input radar image to construct a mapping from input to an output image with suppressed ghosts. The discriminator D evaluates the difference (namely the residual multipath ghosts) between the output image and the ground truth image without multipath ghosts. On the one hand, by training G, the image difference is diminished gradually. In other words, the multipath ghosts are suppressed increasingly in the output image of G. On the other hand, D is trained to improve the ability of evaluating the diminishing difference accompanied with multipath ghosts as much as possible. These two networks G and D fight with each other until G eliminate multipath ghosts. Preliminary simulation results demonstrate GAN can effectively eliminate multipath ghosts in TWRI.