With the continuous evolution of radar RF sensing technology in the Internet of Things (IoT) field, the deployment of co-frequency communication devices within smart home and healthcare environments is becoming increasingly prevalent. Concurrently, the proliferation of RF signals has rendered the challenge of spectrum resource allocation increasingly prominent, exacerbating the issue of severe mutual interference among these co-frequency devices. This severe interference exhibits intricate attributes, characterized by prolonged duration, a broad frequency range, and high level power intensity. When experiencing interference, the target’s echo is notably veiled, rendering recovery through standard technologies exceedingly challenging. From our perspective, in the presence of severe interference, the emphasis ought to be on elegantly reconstructing the target’s echo located in the period of interference, rather than merely eliminating the negative impact on interferences from the raw radar signals. Based on this thought, we have explored the feasibility of a deep network model for interference mitigation of radar signals in this paper, and introduced an interference mitigation method, namely Ra-SPD. The Ra-SPD is crafted through a dual design methodology, integrating the mask-guided spectral decomposition mechanism alongside the region-aware spatial decomposition scheme. The primary objective of the initial design is to enhance the capabilities of the standard deep model, enabling it to distinguish the duration of interference present in radar signals, particularly when the semantic alignment between the restored radar signal and the corresponding ground truth is at risk of being disrupted. Meanwhile, the improvement of local information related to the restored radar signal is accomplished using the spatial decomposition scheme, aimed at increasing the robustness of low-intensity signal segments against severe interference. Experimental results demonstrate superior performance of Ra-SPD at 15%, 40%, and 80% three different signal-to-interference ratio conditions, achieving PSNR: 35.73/32.99/30.28 dB; SSIM: 0.96/0.95/0.92; FID: 23.41/30.97/43.92; and MAE: 0.014/0.024/0.036 across conditions. Our method consistently outperforms six benchmark algorithms in all objective metrics and subjective evaluations, exhibiting 7.6%, 1.5%, 5.5%, and 17.2% average improvements in four metrics relative to the suboptimal method, highlighting significant advantages in interference mitigation and signal quality preservation.

In existing performance analysis of low-Earth orbit (LEO) satellite constellations, the poisson point process (PPP) is always used to model the satellite’s distribution. However, in the actual LEO satellite constellations, satellites are incompletely random like PPP and exhibit a high degree of correlation with each other. In this article, to balance simplicity and accuracy, a tractable model based on spatial repulsion of LEO satellite constellations is derived to capture the correlation of satellites locations, which models a constraint angle$(\phi)$for each satellite, and the coverage performance of the constellations is analyzed via stochastic geometry under shadowed-Rician fading. Specifically, the correlation between the location of satellites is modeled by a constraint angle to compensate for the performance mismatch based on the inherent nonrandom satellites distribution. Furthermore, the distance distribution of users to the satellite is derived under the spherical spatial repulsion model since the actual locations of satellites have a certain degree of regularity among them. In addition, the shadowed-Rician channel is used to model the propagation characteristics of satellite communication channels according to the shadowing condition of the signal on the propagation path. The results show that this model can closely capture the deployment of actual LEO satellite constellations, such as Walker-delta in terms of coverage probability. Overall, the work facilitates general performance analysis and planning of satellite constellations without network-specific orbital simulations and is time-consuming in space.

As one of the crucial sensors for environment sensing, frequency modulated continuous wave (FMCW) radars are widely used in modern vehicles for driving assistance/autonomous driving. However, the limited frequency bandwidth and the increasing number of equipped radar sensors would inevitably cause mutual interference, degrading target detection and producing safety hazards. In this paper, a deep learning-based interference mitigation (IM) approach is proposed for FMCW radars by using the dilated convolution for network construction and a designated contrast learning strategy for training. The dilated convolution enlarges the receptive field of the neural network, and the designated contrastive learning strategy enforces to distinguish better between interferences and desired signals. The results of numerical simulation and experimental data processing show that the dilated convolution-based IM network, compared to the traditional convolution-based ones, can achieve a higher Signal-to-Interference-plus-Noise-Ratio (SINR) and target detection rate. Moreover, the designated contrastive learning strategy enables a better and more stable IM performance without increasing the complexity of the network, which can facilitate faster signal processing.

In this paper, the interference mitigation problem is tackled as a regression problem. A prior-guided deep learning (DL) based interference mitigation approach is proposed for frequency modulated continuous wave (FMCW) radars. Considering the complex-valued nature of radar signals, complex-valued convolutional neural network, which is different from the conventional real-valued counterparts, is utilized as an architecture for implementation. Meanwhile, as the desired beat signals of FMCW radars and interferences exhibit different distributions in the time-frequency domain, this prior feature is exploited as a regularization term to avoid overfitting of the learned representation. The effectiveness and accuracy of our proposed complex-valued fully convolutional network (CV-FCN) based interference mitigation approach are verified and analyzed through both simulated and measured radar signals. Compared with the real-valued counterparts, the CV-FCN shows a better interference mitigation performance with a potential of half memory reduction in low Signal to Interference plus Noise Ratio (SINR) scenarios. The average SINR of interfered signals has been improved from -9.13 dB to 10.46 dB. Moreover, the CV-FCN trained using only simulated data can be directly utilized for interference mitigation in various measured radar signals and shows a superior generalization capability. Furthermore, by incorporating the prior feature, the CV-FCN trained on only 1/8 of the full data achieves comparable performance as that on the full dataset in low SINR scenarios, and the training procedure converges faster.

Human activity recognition (HAR) plays a vital role in many applications, such as surveillance, in-home monitoring, and health care. Portable radar sensor has been increasingly used in HAR systems in combination with deep learning (DL). However, it is both difficult and time-consuming to obtain a large-scale radar dataset with reliable labels. Insufficient labeled data often limit the generalization of DL models. As a result, the performance of DL models will drop when being applied to a new scenario. In this sense, only labeling a small portion of data in the large-scale radar dataset is more feasible. In this article, we propose a semisupervised transfer learning (TL) algorithm, “joint domain and semantic transfer learning(JDS-TL),” for radar-based HAR, which is composed of two modules: unsupervised domain adaptation (DA) and supervised semantic transfer. By employing a sparsely labeled dataset to train the HAR model, the proposed method alleviates the need of labeling a significantly large number of radar signals. We adopt a public radar micro-Doppler spectrogram dataset including six human activities to evaluateJDS-TL. Experiments show that the proposedJDS-TLis able to recognize the six activities with an average accuracy of 87.6% when there are only 10% instances labeled in the training dataset. Ablation analysis also demonstrates the efficiency of the DA and the semantic transfer modules.

Exploiting multiple-input multiple-output (MIMO) technology in UAV networks will mine the potential of spatial multiplexing. However, considering UAVs 3D distribution and line-of-sight (LOS)/non-line-of-sight (NLOS) channel dynamics, multi-antenna transmission will exacerbate network irregularity and render interference management a complicated problem to solve. This paper proposes 3D coordination model for interference management via multi-cell beamforming and signal-level cooperation in multi-antenna UAV networks, and analyzes system performance by deriving semi-closed expression of coverage probability using stochastic geometry. Specifically, UAVs are deployed according to 3D Poisson point process (PPP) with maximum height limit$L$, and user-centric cell group selection is considered according to specific signal threshold$\eta $for interference coordination with the consideration of LOS/NLOS channel conditions. Zero-forcing beamforming and joint transmission is performed within the UAV cluster for interference mitigation with limited channel state information feedback taken account. In addition, considering the irregularity of UAV group, system performance is construed as three scenarios and coverage probability is derived by Gamma approximation with proposed novel coordinate system transformation. Numerical results match well with simulations and provide optimal deployment parameters for UAV deployments, where coverage probability can reach 92% when SIR threshold$T=0$dB.

In multiple unmanned aerial vehicles (UAVs)-aided networks, the time-varying channel caused by the impact of the line of sight (LOS) and nonline of sight (NLOS) leads to signal fluctuations; moreover, the UAV's agility and the constraint of the spatial distribution of blind areas on the UAV's deployment result in interference topology fluctuations. Accordingly, the instantaneous signal-to-interference ratio is fluctuant and cannot accurately reflect network performance. Considering a multi-UAV-aided network with UAV spatial dependence, this article derives the semiclosed expression of long-term average throughput considering the fluctuations of the channel and topology, which are measured by calculating correlation coefficients of the signal and interference. Specifically, the time-varying channel is characterized by incorporating the link-state conversions and randomness of small-scale fading into channel gains, the time-varying network topology is quantified by considering the effect of distance variation caused by mobility on interference topology. In addition, considering multiple UAVs are clustered in corresponding blind areas, UAV networks are modeled as matern cluster processes (MCP), and time-averaged performances at intracluster and extracluster are analyzed, respectively. The numerical results show the effects of clustering parameters and deployment parameters on temporal correlation and long-term average network performance.

The performance of deep learning (DL) algorithms for radar-based human motion recognition (HMR) is hindered by the diversity and volume of the available training data. In this article, to tackle the issue of insufficient training data for HMR, we propose an instance-based transfer learning (ITL) method with limited radar micro-Doppler (MD) signatures, alleviating the burden of collecting and annotating a large number of radar samples. ITL is a unique algorithm that consists of three interconnected parts, including DL model pretraining, correlated source data selection, and adaptive collaborative fine-tuning (FT). Any of the three components cannot be excluded; otherwise, the performance of the entire algorithm decreases. The experiments with a radar data set of six human motions show that ITL achieves state-of-the-art performance for HMR with limited training samples, outperforming several existing transfer learning approaches. Especially, when there are only 100 samples per person per class, ITL yields an F1 score of 96.7%. Last but not least, ITL is more generalized to human motion differences. Though adapted to recognize the persons’ motions in a small-scale target data set, ITL can also classify the persons’ motion data used for pretraining, achieving up to 11.0% F1 score enhancement over the conventional FT method.

Concerns have been raised about our recently published article on the fixed region beamforming using frequency diverse subarray for secure mmWave wireless communications. In a comment, the authors thought our precoding vector normalization method of the sidelobe randomization scheme has a flaw and proposed a non-physical-layer-security-oriented (non-PLS-oriented) normalization method by keeping the norm of the steering vector as a unit. However, we believe our PLS-oriented normalization method of the transmit beamforming vector is correct and reasonable from the PLS perspective, i.e., we hope to keep the target use's beampattern gain unit. In this rebuttal, we further clarify and justify our scheme to show its correctness. In addition, we also present a generalized normalization method to compare our proposed PLS-oriented scheme and the non-PLS-oriented scheme in the comment to offer useful insights.

Automatic modulation classification facilitates many important signal processing applications. Recently, deep learning models have been adopted in modulation recognition, which outperform traditional machine learning techniques based on hand-crafted features. However, automatic modulation classification is still challenging due to the following reasons. Existing deep learning methods are only applicable to the data of the same distribution. In practical scenarios, data distribution is varying with sampling frequency, thus domains with different sampling rates are formed. Besides, it is difficult to construct large-scale well-annotated datasets for all domains of interest. We define the domain with sufficient data as the source domain, while the domain with insufficient data as the target domain. Obviously, the classification model performs weakly in the target domain. To address these challenges, we propose an adversarial transfer learning architecture (ATLA), incorporating adversarial training and knowledge transfer in a unified way. Adversarial training performs an asymmetric mapping between domains and reduces the domain shift. Knowledge transfer is used to mine prior knowledge from the source domain. Experimental results demonstrate that the proposed ATLA substantially boosts the performance of the target model, which outperforms the existing parameter-transfer approach. With half of the training data reduced, the target model achieves competitive recognition accuracy to supervised learning. With one-tenth of training data, the promoted accuracy is up to 17.3% points.

In remote sensing, micro-Doppler signatures are widely used in moving target detection and automatic target recognition. However, since Doppler signatures are easily affected by the moving direction of the target, prior information of aspect angle is essential for spectral analysis. Thus, a micro-Doppler-based classifier is considered to be “angle-sensitive.” In this article, we propose an angle-insensitive classifier for the omnidirectional classification problem using the monostatic radar through a proposed new convolutional neural network. We further provide a sensible definition of “angle sensitivity,” and perform experiments on two data sets obtained through simulations and measurements. The results demonstrate that the proposed algorithm outperforms both feature-based and existing deep-learning-based counterparts, and resolve the issue of angle sensitivity in micro-Doppler-based classification.

Millimeter-wave (mmWave) using conventional phased array (CPA) enables highly directional and fixed angular beamforming (FAB), therefore enhancing physical layer security (PLS) in the angular domain. However, as the eavesdropper is located in the direction pointed by the mainlobe of the information-carrying beam, information leakage is inevitable and FAB cannot guarantee PLS performance. To address this threat, we propose a novel fixed region beamforming (FRB) by employing a frequency diverse subarray (FDSA) architecture to enhance the PLS performance for mmWave communications. In particular, we carefully introduce multiple frequency offset increments (FOIs) across subarrays to achieve a sophisticated beampattern synthesis that ensures a confined information transmission only within the desired angle-range region (DARR) in close vicinity of the target user. More specifically, we formulate the secrecy rate maximization problem with FRB over possible subarray FOIs, and consider two cases of interests, i.e., without/with the location information of eavesdropping, both turn out to be NP-hard. For the unknown eavesdropping location case, we propose a seeker optimization algorithm to minimize the maximum sidelobe peak of the beampattern outside the DARR. As for the known eavesdropping location case, a block coordinate descend linear approximation algorithm is proposed to minimize the sidelobe level in the eavesdropping region. Moreover, we propose an inverted subarray subset technique to further randomize the sidelobes against sensitive eavesdropping. By using the proposed FRB, the mainlobes of all subarrays are constructively superimposed in the DARR while the sidelobes are destructively overlayed outside the DARR. Therefore, FRB exhibits prominent effect on confining information transmission within the DARR. Numerical simulations demonstrate that the proposed FDSA-based FRB can provide superior PLS performance over the CPA-based FAB.

Wearable activity recognition is an important area for healthcare applications, especially for children and elderly people. Current research proves that deep learning techniques are capable of learning deep features from the sensor raw signal. However, this task faces two challenges. First, it is hard to capture the dynamic and non-linear interactions between different sensor modalities and time slots. Besides, the spatial relationship between sensors convey significant information for wearable activity recognition, thus how to model the sensor relationship is one of the problems that need to be solved. In this work, we propose GraphConvLSTM, a scalable framework that model spatial relationship, local interaction, and long-term temporal dependency simultaneously. We evaluate our model on two real-world activity recognition datasets, and it achieves 1.75%, 2.68% accuracy improvements over existing methods (accuracy: 90.83%, 93.08%) for wearable activity recognition task, which demonstrates the ability of the proposed approach to learn spatial, local and global temporal features.

In a smart home environment, assisted living has been a topic of great research over the past decade. Human motion analysis is considered as a key technology for living states recognition in an assisted living system. Recent research has proved that rich information can be obtained from human movements, such as the motion category, moving patterns, and human identity. In this paper, a nonintrusive human movement sensing system is established with a mono-static ultrawide bandwidth radar. Then, we propose a well-designed joint motion classification (MCL) and person identification (PID) convolutional neural network (named as “JMI-CNN”). To recognize human motions and identities simultaneously, the network employs a multitask learning scheme as well as the attention mechanism and the hierarchical feature reuse strategies. We report the experimental result on the data from 15 individuals, each performing six motions. It shows that the model achieves a promising performance of 80.57% on the joint task, while the accuracy for MCL and PID are 98.50% and 80.92%, respectively. Moreover, we carry out ablation studies to evaluate the design principles of the proposed method. Discussions on the impact of signal noise ratio and slow-time window length are also conducted.

Human target detection and tracking have great potential in military, safety, security and entertainment applications. In this paper, a complete processing procedure is proposed for human tracking in foliage-penetration environment by multistatic radar. It consists of five main steps, including clutter suppression, target detection, measurement estimation, target localization and target tracking. Exponential average background subtraction is applied for clutter suppression. Ordered statistics constant false alarm rate (OS-CFAR) detector is used to detect targets in the range profile. The range measurement estimation is realized by a window filter. The S-D assignment algorithm is introduced for multi-target localization. Target tracking is realized by a Kalman filter based multi-target tracking (MTT) system. The experimental results verify the effectiveness of the proposed human tracking procedure.