Real-time positioning feature is becoming an essential part in a variety of military and civilian applications. In particular, the cooperation of multisensors helps to improve navigation accuracy and the universality of the system. In this article, we propose a novel robust fault-tolerant federated filter based on strapdown inertial navigation system (SINS), global navigation satellite system (GNSS), celestial navigation system (CNS), and doppler velocity log (DVL). In the proposed algorithm, the position information of the GNSS, the velocity information of the DVL, and the attitude information of the CNS are introduced as measurement information to correct the divergence error of SINS. The federated filter is exploited for its flexibility and capability in fault toleration. To amend fault and recover information in local filters, a simplified state Chi-square test (SSCST) is utilized as the fault detection, isolation, and recovery procedure. Meanwhile, a new adaptive information sharing factor algorithm based on the results of SSCST and the residuals between the actual observations and the predicted observations is designed, which can adaptively reflect the performance of each local filter. Experimental results show that the accuracy of the proposed integrated navigation and positioning algorithm can be improved and the system remains stable even when the error occurs.

Stacked intelligent metasurfaces (SIM) capable of beamforming in the wave domain is a promising technology for realizing low-cost and high-efficiency integrated sensing and communication (ISAC) systems. In this paper, we investigate the trade-off design in SIM-aided ISAC systems with discrete phase shifts. Specifically, we propose an interference management method to improve the multiuser communication quality and a beam gain approximation method to enhance the sensing beam gain. The resulting power allocation and phase-shift optimization problems are intractable discrete non-convex problems. Therefore, we develop an efficient alternating optimization algorithm to decouple the complicated problem and then propose an efficient variable separation approach to obtain closed-form solutions for all subproblems. Simulation results show that the proposed discrete phase-shifted design achieves a 40% performance improvement compared to the quantization of the continuous phase-shifted design in existing studies. Moreover, the proposed method is able to achieve better target sensing and more effective multiuser interference management in real discrete systems compared to continuous phase-shift quantization.

Cell-free massive multiple-input multiple-output (MIMO) network includes numerous geographically distributed access points (APs) serving users through coherent transmission and reception. To achieve scalability, each user should be assigned a personalized cluster of APs. In this letter, we propose a deep reinforcement learning (DRL)-based approach to determine the cluster of APs for each user while satisfying constraints on minimum rates for all users, considering practical concerns such as pilot contamination and statistical channel state information (CSI). Simulation results demonstrate that the proposed DRL-based AP selection scheme outperforms other conventional schemes.

The baleful jamming attacks are destructive to the global navigation satellite system (GNSS) receivers. Detecting the jamming type accurately at a GNSS receiver facilitates to select the corresponding mitigation method. Unfortunately, the recognition method based on the manual selection of the signal features relies on the independence of features. Considering the information of similar interference signals in the frequency domain is comparable, the existing deep learning methods such as convolutional neural networks (CNN) with local receptive fields have difficulty in modeling and processing the global correlation between signals. To this end, a dual graph convolution network (GCN) with adaptive weight learning (AWL-DGCN) method is proposed in this paper. The interference signals are converted into graph data, and the GCN, which has demonstrated advantages in processing graph data, is employed for their analysis. More specifically, three kinds of time and frequency domain features are extracted to establish different homogeneous graphs between signals. Furthermore, the current graph aggregation method simply adds the homogeneous graphs which contain different information, making it difficult to dynamically balance the contributions of different information, which affects the effectiveness of message passing. Therefore, a dual GCN structure is proposed for adjusting adjacency matrix weights adaptively, which is composed of the weight learning module and the classification module. The experimental results on simulation, public and real-world dataset indicate that the proposed method possesses lower complexity, shorter reasoning time and higher recognition accuracy, which holds potential for application in scenarios where a timely response to GNSS interference attacks is necessary.

Recently, quantization has been an effective technique for large-scale image retrieval, which can encode feature vectors into compact codes. However, it is still a great challenge to improve the discriminative capability of codewords while minimizing the quantization error. This letter proposes Dual Distance Optimized Deep Quantization (D ² ODQ) to deal with this issue, by minimizing the Euclidean distance between samples and codewords, and maximizing the minimum cosine distance between codewords. To generate the evenly distributed codebook, we find the general solution for the upper bound of the minimum cosine distance between codewords. Moreover, scaler constrained semantics-preserving loss is considered to avoid trivial quantization boundary, and ensure that a codeword can only quantize the features of one category. In contrast to state-of-the-art methods, our method has a better performance on three benchmark datasets.