Frequency shift jamming (FSJ) significantly challenges the detection performance of radar with the linear frequency modulation (LFM) waveform, since jamming signals keep LFM coherence with only frequency shift modulation. In this paper, an adaptive time-frequency filtering algorithm is developed, which exploits the different time-frequency relationships between FSJ signals and true targets to construct the adaptive suppression filters. The difference in the filtering energy loss ratio is measured by clustering technique to determine the FSJ targets. Extensive simulation experiments show that the proposal performs effectively and stably for FSJ suppression.

With the development of space resources, high-precision positioning is a significant way to obtain the on-orbit status information of space targets. Ground radar interferometric is an active space target positioning method, which can provide positioning information in all-weather and all-time. However, the directly measured interferometric phase will be ambiguous when the range difference between space targets and radar sites exceeds a certain boundary. In this letter, phase ambiguity resolution (AR) is established as a hidden Markov model (HMM) dynamic optimization problem in combination with the space target's movement process at different pulse times, in which no auxiliary baseline is required compared with traditional methods. And the Viterbi algorithm is used to solve the optimal ambiguous number (AN) sequence. Since the sequence decoding is mainly based on solving the optimal AN path that to explain the target position at different times best, a transmission technique of staggered carrier frequency signals is adopted, which significantly increases the robustness of phase AR. Extensive simulation experiments show that the proposal performs effectively and accurately for high-precision space target positioning.

The unfolding network, which combines traditional compressive sensing (CS) methods and deep learning, has been widely used in sparse aperture (SA) inverse synthetic aperture radar (ISAR) imaging because of its good interpretability, high reconstruction accuracy, and computational efficiency. However, most unfolding networks view SA ISAR imaging under different sampling matrices as diverse problems, resulting in poor network transferability when the sampling matrix changes. In this article, a complex-valued residual alternating direction method of multipliers-Net (CV-Res-ADMM-Net) is designed to address this issue. Based on the radar signal's complex-valued (CV) characteristics, we designed a residual learning-based CV convolutional neural network (CNN) that includes three modules: feature extraction, nonlinear mapping, and feature accumulation to implement the nonlinear operation to enhance the network's ISAR image feature extraction capability. The residual learning technique can accelerate the training process and smooth network information transmission. Moreover, a multi-sampling matrix generation approach is developed, which can improve the training diversity and maintain more structural information on ISAR images for more satisfactory reconstruction. Simulated and real-measured experiments verify that CV-Res-ADMM-Net has good transferability when the sampling matrix changes. Experimental results show the proposal can achieve fast SA ISAR imaging of maneuvering targets.

By unfolding traditional optimization algorithms into the form of neural networks, the unfolding network methods have attracted more and more attention in sparse inverse synthetic aperture radar (ISAR) imaging because of their high reconstruction performance and good interpretability. However, existing unfolding network methods mainly focus on one-dimensional sparse ISAR imaging and cannot be directly applied to two-dimensional (2D) sparse ISAR data. For this reason, a novel learned 2D-two-step iterative shrinkage/thresholding algorithm (L-2D-TwISTA) is proposed for high efficiency and high accuracy 2D sparse ISAR imaging. Specifically, each stage of L-2D-TwISTA corresponds to an iterative solution step of the developed 2D-TwISTA approach. Moreover, a complex-valued residual network is designed in L-2D-TwISTA to improve training efficiency and solve the nonlinear problem of proximal mapping of the 2D-TwISTA more effectively. The experimental results of real-measured data confirm that the L-2D-TwISTA can realize high-performance 2D sparse ISAR imaging.

Recent research has shown that the unrolling network is an effective and efficient method for suppressing interrupted sampling repeater jamming (ISRJ). However, in practice, variability in jamming location and duty cycle often prevents most unrolling networks from achieving satisfactory ISRJ suppression performance. To address this challenge, this article introduces complex-valued fast iterative shrinkage-thresholding algorithm network (CV-FISTA-Net), a complex-valued unrolling network inspired by the FISTA method. CV-FISTA-Net replaces the conventional soft-thresholding operator with a complex-valued convolutional neural network (CV-CNN) and a residual learning approach. This enhances training efficiency, signal denoising, and reconstruction fidelity. Furthermore, a random projection augmentation strategy is introduced to diversify the training set, bolstering the network's robustness against variations in ISRJ location and duty cycle. To improve generalization, a transfer learning technique is integrated, ensuring strong jamming suppression performance even with limited training data or under changing conditions, such as varying signal-to-noise ratios. Leveraging the computational efficiency of CV-FISTA-Net, we also propose an effective strategy for joint ISRJ suppression and inverse synthetic aperture radar (ISAR) imaging. Experiments on both simulated and real-measured data validate that CV-FISTA-Net achieves precise and robust ISRJ suppression amidst dynamic changes in jamming location and duty cycle.