Space targets move in orbit at a very high speed, so in order to obtain high-quality imaging, high-speed motion compensation (HSMC) and translational motion compensation (TMC) are required. HSMC and TMC are usually adjacent, and the residual error of HSMC will reduce the accuracy of TMC. At the same time, under the condition of low signal-to-noise ratio (SNR), the accuracy of HSMC and TMC will also decrease, which brings challenges to high-quality ISAR imaging. Therefore, this paper proposes a joint ISAR motion compensation algorithm based on entropy minimization under low-SNR conditions. Firstly, the motion of the space target is analyzed, and the echo signal model is obtained. Then, the motion of the space target is modeled as a high-order polynomial, and a parameterized joint compensation model of high-speed motion and translational motion is established. Finally, taking the image entropy after joint motion compensation as the objective function, the red-tailed hawk–Nelder–Mead (RTH-NM) algorithm is used to estimate the target motion parameters, and the joint compensation is carried out. The experimental results of simulation data and real data verify the effectiveness and robustness of the proposed algorithm.

When a space target malfunctions and is no longer controlled by its attitude control system, it usually tumbles in orbit and exhibits a slow spinning state. Accurately estimating the on-orbit attitude of spinning space targets is of vital importance for ensuring the operation of space assets. Moreover, it plays a significant role in tasks such as reentry observation and collision avoidance. Currently, most existing methods estimate the attitude of space targets by using a single inverse synthetic aperture radar (ISAR) for long-term observation. However, this approach not only requires a long observation time but also fails to estimate the attitude of spinning targets. To address these limitations, this paper proposes a novel approach for estimating the attitude of spinning space targets, which utilizes the joint observations of a multiple-station ISAR. Specifically, the proposed method fully exploits the projection principle of ISAR imaging and uses an ISAR high-resolution network (ISAR-HRNet) to automatically extract the projection features of typical components of the target. Then, the analytical expressions for the target’s instantaneous attitude and spin vector under the multi-station observation imaging projection model are derived. Based on the extracted features of the typical components, the lengths, orientations, and spin vectors of the space target are determined. Importantly, the proposed method can achieve the attitude estimation of the spinning space targets within a single observation period, without the need for manual intervention or prior information about the target’s three-dimensional (3D) model. Additionally, the analytical method for solving the spin vector offers high efficiency and accuracy. Finally, the effectiveness of the proposed attitude estimation algorithm is verified by experiments on simulated data, and the performance of the ISAR-HRNet is also tested in the key point extraction experiments using measured data.

Semantic segmentation tasks encompass various applications, such as autonomous driving, medical imaging, and robotics. Achieving accurate semantic information retrieval under conditions of high dynamic range and rapid scene changes remains a significant challenge for image-based algorithms. This challenge is primarily attributable to the limitations of conventional image sensors, which can experience motion blur or exposure artifacts. In contrast, event-based vision sensors, which asynchronously report changes in pixel intensity, offer a compelling solution by acquiring visual information at the same rate as the scene dynamics, thereby mitigating these limitations. However, we encounter a significant challenge in event-based semantic segmentation tasks: the need to expend time on converting event data into frame images to align with existing image-based semantic segmentation techniques. This approach squanders the inherently high temporal resolution of event data, compromising the accuracy and real-time performance of semantic segmentation tasks. To address these issues, this work explores a sparse semantic segmentation approach that directly addresses event data. We propose a network named EventSegNet that improves the ability to extract geometric features from event data by combining geometric feature enhancement operations and attention mechanisms. Based on this, we propose a large-scale event-based semantic segmentation dataset that provides labels for each event. Our approach achieved a new F1 score of 84.2% on the dataset. In addition, a lightweight and edge-oriented AI inference deployment technique was implemented for the network model. Compared to the baseline model, the optimized network model reduces the F1 score by 1.1% but is more than twice as fast computationally, enabling real-time inference on the NVIDIA AGX Xavier.

Spiking neural networks (SNNs), a class of brain-inspired computational models, are characterized by asynchrony, discreteness, and sparseness, attributes that are congruent with neuromorphic data. Recent research has been dedicated to applying SNNs to neuromorphic data classification tasks for enhancing classification performance. However, this task is significantly constrained by the expressive abilities of neurons and the efficient feature extraction technology for spiking signals. Motivated by these, we first propose the generalized parameter leaky integrate-and-fire (G-PLIF) neurons model, in which neuron membrane potential threshold is variable to simulate neurons with different biological characteristics. In G-PLIF, the neuron membrane potential threshold is constantly changing during the training process, thereby enhancing the dynamic heterogeneity and representational ability of neurons. Then, based on G-PLIF, we designed a Multi-Scale-Residual Spiking Neural Network (MSR-SNN) specifically for the classification tasks of neuromorphic vision objects. The concept of a multi-scale pyramid has been integrated into the residual blocks of the network, fortifying its ability to extract deep-level, multi-dimensional features. The proposed method is evaluated on challenging neuromorphic datasets DVS-CIFAR10 and DVS-Lip, and extensive experimental results demonstrate that our method achieved state-of-the-art performance.

With the help of specific emitter identification (SEI), the control efficiency of the satellite communication systems can be effectively improved by discriminating the individual satellite. In recent years, deep learning has been introduced into SEI to enhance identification performance because of its powerful classification capability. However, classical real-valued neural networks exhibit some limitations in extracting the radio frequency fingerprint (RFF) features from complex signals, limiting the improvement of identification accuracy. Thus, we proposed a complex-valued conditional adversarial generative network (CC-GAN) which can directly deal with complex signals. Through adversarial learning between the generator and the discriminator, the generator implements direct mapping from the dynamic noisy signals to the noise-free signals. In addition, an auxiliary classifier is introduced into the discriminator to make the discriminator able to label the sample, which effectively compress the proposed model. The experimental results for a signal dataset collected in a real environment demonstrated that the proposed model is superior to the traditional denoising methods in denoising performance, which effectively improves the identification accuracy under dynamic noises. Furthermore, the proposed model outperforms other deep learning models in terms of identification performance under various SNRs, which can effectively improve the robustness and adaptability of the SEI system for communication satellites in dynamic noisy environments.