In view of the practical value of open datasets of hyperspectral images (HSIs), HSI open-set classification (OSC) has attracted more and more attention. Existing HSI OSC methods are usually based on learning labeled samples to identify unknown classes. However, due to the complex high-dimensional characteristics of HSIs and the limited number of labeled samples, the recognition of unknown classes based only on limited labeled samples often has low and unstable accuracy. To address this problem, we propose a semi-supervised graph constraint dual classifier network (SSGCDCN) that can achieve efficient and stable OSC by learning unknown class features and relationships among samples. First, a dual classifier consisting of a multi-classifier and multiple binary classifiers is constructed, which has the ability to discover the unknown class samples by assigning and enabling pseudo-labels to participate in model training to achieve unknown class feature learning. Then, to improve the classification accuracy of both known and unknown classes, a homogeneous graph constraint is imposed on SSGCDCN to learn the relationship information among samples (including labeled and unlabeled samples). This constraint can bring the features of similar samples closer while pushing apart features of dissimilar samples. Experiments evaluated on three datasets demonstrate that the proposed method can obtain superior OSC performance than other state-of-the-art classification methods.

Hyperspectral image (HSI) captures hundreds of narrow bands, offering abundant spatial and spectral information across various application domains. In recent years, deep learning techniques have shown strong potential in HSI processing, with a range of models being explored. However, the fixed geometric structure of convolutional kernels in CNN model impedes interactions between long-term dependencies, and existing transformer methods have not fully harnessed the spectral structural characteristics of HSI data. To address these issues, we proposed a multiscale samples transformer (MSST) that effectively extracts spatial and spectral features via two separate branches respectively. In spectral branch, the spectral sequence of the individual target pixel is used as input sample, a spectral transformer (SpecFormer) is proposed to comprehensively extract the spectral structural features. In spatial branch, three-dimensional tensor consisting of target pixel and their neighboring pixels is used as input sample, dimension reduction is performed, and a spatial transformer (SpatFormer) is designed to efficiently extract the deep spatial features of HSI data. Consequently, MSST enlarges the receptive field and captures the comprehensive joint spatial-spectral features. Extensive experiments conducted on five widely used HSI datasets demonstrate the superior classification performance of the proposed framework. Our source codes will be available at https://github. com/zhwt-xidian/Multi-Scale-Samples-Transformer_MSST.

Noninvasive fetal ECG (FECG) monitoring holds significant importance in ensuring the normal development of the fetus. Since FECG is usually submerged by maternal ECG (MECG) and background noise in abdominal ECG (AECG), it is challenging to exactly restore the waveform details of FECG from AECG. To address this issue, a temporal convolutional generative adversarial network (TCGAN) is proposed for FECG extraction using single-channel AECG. In order to utilize both the global and local ECG features in time domain, we built an encoder-decoder architecture for designing generator. The model architecture consists of temporal convolution blocks, transpose convolutions and skip connections. The skip connections attempt to achieve the purpose of amalgamating information from feature maps extracted by convolutional layers using transpose convolution operations, which facilitates the decoder for extracting more detail information. TCGAN is rigorously evaluated using both synthetic dataset FECGSYDB and real-world dataset ADFECGDB. The experimental results on above datasets demonstrate the outstanding performance of TCGAN in terms of fetal QRS complex detection, achieving PPV of 99.54% and 99.02%, respectively. Comparing with the state-of-the-art methods, TCGAN could extract FECG with well-preserved waveform details. This helps doctors achieve more accurate assessment of fetal development.

As an essential task in remote sensing, hyperspectral image classification (HSIC) has been extensively studied. Although many methods based on deep learning have been inducted to improve the performance of HSIC, they still face two challenges: 1) Can the joint spectral–spatial information in the hyperspectral image (HSI) be effectively utilized? 2) Does each pixel in a certain sample contribute equally to classification? We exploit a spectral–spatial attention tensor network (SSATN) in this letter, where the coordinate attention (CA) mechanism is first introduced into a full tensor network. We present a learnable tensor squeezing projection layer (TSPL) instead of the classical pooling layer for the CA block, which enables the network to selectively focus on discriminative features in spectral and spatial. Besides, the SSATN accepts raw HSI data as the input without dimensionality reduction in advance. Compared with the state-of-the-art convolutional neural network (CNN) methods, SSATN can effectively capture certain spectral–spatial features via tensor transformation with fewer parameters. The experimental results on two widely used HSI datasets prove the advantage of the SSATN method.

In wideband DOA estimation scenario, half-wavelength condition for a uniform linear array (ULA) is usually not satisfied in entire frequency bands, which may result in spatial aliasing in higher frequency bands. Especially when the strong and weak sources exist simultaneously, the aliasing components from the strong sources almost always lead to failure detection of weak sources. To handle this issue, the wideband spatial aliasing model for each sub-band in uniform linear arrays (ULAs) is built. By exploiting the Cauchy-Schwarz inequality, we proposed a theorem for determining the lower bound of the time-averaged energy-density spectrum without spatial aliasing. Based on the derived theorem and the similar structure of the spectrum between adjacent sub-bands, three unambiguous wideband DOA estimation schemes are proposed. Firstly, two unambiguous spectra are derived by applying the proposed theorem to conventional beamforming spectrum and Capon spectrum respectively. The former has better aliasing suppression performance than the latter, but with lower resolution. By incorporating the advantages of themselves, a quadratically constrained beamforming (QCB) method is proposed, where an elegant closed form expression for the output power is derived for DOA estimation. Compared to the frequency-difference (FD) approaches and multi-stage approach, the proposed method has higher spatial resolution and detection probability for weak sources. Both simulation and experimental results are provided to demonstrate the effectiveness of the proposed method.

Nowadays, foundation models have demonstrated exceptional performance across numerous downstream tasks. However, the effective application of these models to hyperspectral image classification (HSIC) is challenged by the unique characteristics of hyperspectral data, including high dimensionality, high variability, and high spatial structure complexity. Therefore, methods need to be developed, which leverage the advantages of foundation models while addressing these challenges. First, a novel HSIC algorithm based on a frozen-parameter segment anything model (SAM) encoder, called SAGFFNet, is proposed. This framework represents the first attempt to use a frozen SAM encoder for global feature extraction and to use spectral dimension data as prompts, enabling precise global spatial-spectral feature extraction with the aid of spectral information. Second, by introducing the self-adaptive padding mechanism and the global feature extraction subnetwork (GFEsNet), the model is enabled to extract distinctive and discriminative features for each category from hyperspectral data through varying padding sizes, thereby enhancing the feature extraction and generalization capabilities of the foundation model. Subsequently, the spectral feature prompt subnetwork (SFPsNet) is designed to extract spectral feature information from samples of different classes as prompt features, assisting the framework in better understanding the global features extracted by GFEsNet. Finally, the semantic information decoder subnetwork (SIDsNet) is introduced as a semantic information decoder, achieving efficient fusion of global spatial-spectral features and spectral prompt features, which significantly improves classification performance. Experiments conducted on four hyperspectral image datasets show that the proposed method outperforms nine existing approaches in terms of classification accuracy.

Hyperspectral image (HSI) is widely used in real-world classification tasks since it contains rich spatial and spectral features consisting of hundreds of continuous bands. However, the rich spectral features always result in “dimension disaster” for HSI classification. Some traditional methods adopt the strategy of dimensionality reduction in advance to alleviate the influences of high spectral feature dimension, which usually results in degraded classification accuracy due to the hard coupling of the dimensionality reduction model and the classifier. In this letter, we propose a full tensor network for HSI classification based on Tucker decomposition-based hidden layer and CP decomposition-based classification layer. The proposed network can efficiently extract the high-level structural features directly from the original data and classify HSI in a single network without dimension reduction procedure in advance. Moreover, the proposed model has a simple structure and holds less parameters than the state-of-art classification methods. The experimental results on Pavia University and Salinas datasets show that for training ratio of 5%, the overall accuracy of our model are averaged 3.41% higher than CNN models while the network parameters are averaged 80.44% less, which validate the merits of the proposed method.

In order to track a changing channel in multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) systems, it is a priority to estimate channel impulse response adaptively. In this paper, we propose an adaptive blind channel estimation method based on parallel factor analysis (PARAFAC). We used an exponential window to weigh the past observations; thus, the cost function can be constructed via a weighted least squares criterion. The minimization of the cost function is equivalent to the decomposition of a third-order tensor, which consists of the weighted OFDM data symbols. By preserving the Khatri-Rao product, we used a recursive least squares solution to update the estimated subspace at each time instant, then the channel parameters can be estimated adaptively, and the algorithm achieves superior convergence performance. Simulation results validate the effectiveness of the proposed algorithm.

This paper deals with the nonunitary joint block diagonalization (JBD) of a set of given matrices for nonsquare mixing. As is known that the elimination of the degenerate solutions is a crucial problem in nonunitary JBD. Unlike the existing method, which optimizes a penalty term based least squares criterion to eliminate the degenerate solutions. In this paper we seek a well conditioned solution to nonunitary JBD instead. From this point of view, the nonunitary JBD problem is reformulated as a multicriteria optimization model. The resulting algorithm is numerically robust, and can be used in nonsquare mixing scenario. Moreover, it outperforms the existing JBD algorithms in terms of convergence rate and stability, which has been verified by our simulation results.

This paper addresses the problem of adaptive source extraction via the canonical correlation analysis (CCA) approach. Based on Liu's analysis of CCA approach, we propose a new criterion for source extraction, which is proved to be equivalent to the CCA criterion. Then, a fast and efficient online algorithm using quasi-Newton iteration is developed. The stability of the algorithm is also analyzed using Lyapunov's method, which shows that the proposed algorithm asymptotically converges to the global minimum of the criterion. Simulation results are presented to prove our theoretical analysis and demonstrate the merits of the proposed algorithm in terms of convergence speed and successful rate for source extraction.

In this letter, we introduce a fast and computationally efficient iterative algorithm for joint zero diagonalization of a set of complex-valued target matrices. The proposed algorithm is actually a low complexity version of FJZD algorithm, it has a computational complexity of O(K N2), whereKandNare the number and dimension of the target matrices respectively. Moreover, the proposed algorithm is superior to FJZD in terms of interference to signal ratio. Simulation results demonstrate the good performance of the proposed algorithm.

A new iterative algorithm for the nonunitary joint zero diagonalization of a set of matrices is proposed for blind source separation applications. On one hand, since the zero diagonalizer of the proposed algorithm is constructed iteratively by successive multiplications of an invertible matrix, the singular solutions that occur in the existing nonunitary iterative algorithms are naturally avoided. On the other hand, compared to the algebraic method for joint zero diagonalization, the proposed algorithm requires fewer matrices to be zero diagonalized to yield even better performance. The extension of the algorithm to the complex and nonsquare mixing cases is also addressed. Numerical simulations on both synthetic data and blind source separation using time-frequency distributions illustrate the performance of the algorithm and provide a comparison to the leading joint zero diagonalization schemes.