With the recent improvement of deep learning (DL) techniques and computer hardware capabilities, neural networks are widely used to monitor massive sensor data and detect earthquakes in them. This makes designing fast, accurate, and generalized DL models necessary for an active field of research for automatic seismic phase picking. A seismic phase picking network called MFFnet is proposed to fuse power spectral density (PSD), expert knowledge, spectrograms, recurrence plots (RPs), and Gramian angle fields. The network uses fast Fourier convolution (FFC) on 2-D representations to extract more interpretable features. Considering the high proportion of noisy signals in field applications, MFFnet uses focal loss (FL) as the loss function to improve network accuracy. Experimental results show that MFFnet achieves precision, recall, and accuracy with 0.96, 0.98, and 0.98, respectively, in seismic phase detection tasks. Shapley value is used to evaluate the relationship between features and network predictions. Compared with other DL networks, the feature extraction approach used in this letter is more explanatory and provides greater confidence in the results.

Tomographic SAR (TomoSAR) can achieve 3-D imaging for observation targets through tomographic synthetic aperture, and shows good characteristics in urban building information extraction and scene 3-D inversion. Though the existing CS-based imaging algorithms can achieve high-resolution imaging results, requiring multiple iterations and manual adjustment of hyper-parameters. Currently, deep learning techniques in TomoSAR show great advantages in improving the imaging accuracy and efficiency. Inspired by deep unfolding, we unfolded the CS-based ADMM algorithm and mapped it into deep unfolded ADMM-net (DUADMM-net), so as to achieve high-resolution TomoSAR imaging. DUADMM-net consists of reconstructed signal estimation module, nonlinear fitting module and multiplier update module. The introduction of convolutional layers enhances the learning ability and nonlinear fitting ability. Compared to the conventional sparse imaging algorithms, the experimental imaging results and quantitative indicators demonstrate the effectiveness and efficiency of DUADMM-net.

Image resolution is the key point for the 3-D synthetic aperture radar (SAR) application, especially in small-scale scene observation. The traditional filter-based image enhancement algorithms used for 3-D SAR may suffer from quality degeneration in case of parameter mismatch. This paper proposes a robust and efficient convolutional neural network (CNN) based U-net framework for 3-D SAR image enhancement. The U-net extracts image features in down sampling and up sampling, which is realized by max pooling and deconvolution layers. We use the mean square error(MSE) as the loss function to estimate the difference between the predicted images and the label, while Adam optimizer updates parameters to achieve the global minimum MSE. Both simulation and measured data verify the effectiveness of the network. The results demonstrate that the U-net outperform some traditional filter-based algorithms.

The compressed sensing (CS) method is widely utilized in the field of radar sparse imaging. However, it always encounters enormous iterations and low generalizability. To solve these problems, in this paper, we propose a novel learning-based sparse imaging network architecture, i.e., Split Unfolding Sparsity-Driven Network (SSD-Net), for 3D synthetic aperture radar (SAR) imaging. By combining the model-based SAR imaging method and data-driven deep learning method, SSD-Net has favorable explainability and generalization abil-ity to produce 3D SAR images. The deep hierarchical ar-chitecture of SSD-Net is obtained by combiningthe radar nonlinear operator and the split Bregman method. The exper-iments demonstrate that the proposed SSD-Net outperforms other state-of-the-art methods in the field of SAR imaging.

The sparse imaging method based on compressed sensing (CS) is widely used in the field of millimeter-wave (MMW) synthetic aperture radar (SAR) imaging. However, 3D sparse imaging is limited by the difficult parameter tuning, the huge computational load, and the low processing efficiency. In addition, due to the motion errors and model mismatch, it is difficult to obtain well-focused results without error correction techniques. To address these issues, we propose a deep learning framework that integrates 3D sparse imaging and autofocusing, named 3D Sparse Autofocusing Network (SAF-3DNet) for MMW SAR data processing. The network is constructed based on an auto-encoder, which can optimize parameters without effective ground truth. The backbone structure of the encoder is expanded by approximate message-passing (AMP), and the operators in the frequency domain are used to replace the traditional matrix-vector CS model, which avoids large-scale matrix multiplication and other operations, and greatly improves the operation efficiency. In addition, the 2D phase error estimation in the cross-range plane is embedded into the sparse imaging models, enabling simultaneous 3D imaging and autofocusing. The decoder is designed as a mapping from the autofocusing results to the echo data. Experimental results based on both simulated and measured data demonstrate the proposed SAF-3DNet can achieve well-focused 3D reconstruction within an ephemeral time, which expresses the potential of 3D MMW SAR real-time and high-quality imaging.

Next-generation sequencing (NGS) has revolutionized the detection of structural variation in genome. Among NGS strategies, read depth is widely used and paramorphism information contained inside is generally ignored. We develop an algorithm that can fully exploit both read depth and paramorphism information. We embed mutation procedure in our system model for estimating prior likelihood of single nucleotide base. Hidden Markov model (HMM) is used to connect single base into segments and belief propagation algorithm is performed for the optimal solution of the HMM model. Simulations show promising results in detecting important types of structural variation. We have applied the algorithm on the maize B73 and MO17 genome data and compared the results with those obtained from array CGH method based micro-array data. Inconsistency between the two sets of data is discussed.

The knowledge of multiple conformational states is a prerequisite to understand the function of membrane transport proteins. Unfortunately, the determination of detailed atomic structures for all these functionally important conformational states with conventional high-resolution approaches is often difficult and unsuccessful. In some cases, biophysical and biochemical approaches can provide important complementary structural information that can be exploited with the help of advanced computational methods to derive structural models of specific conformational states. In particular, functional and spectroscopic measurements in combination with site-directed mutations constitute one important source of information to obtain these mixed-resolution structural models. A very common problem with this strategy, however, is the difficulty to simultaneously integrate all the information from multiple independent experiments involving different mutations or chemical labels to derive a unique structural model consistent with the data. To resolve this issue, a novel restrained molecular dynamics structural refinement method is developed to simultaneously incorporate multiple experimentally determined constraints (e.g., engineered metal bridges or spin-labels), each treated as an individual molecular fragment with all atomic details. The internal structure of each of the molecular fragments is treated realistically, while there is no interaction between different molecular fragments to avoid unphysical steric clashes. The information from all the molecular fragments is exploited simultaneously to constrain the backbone to refine a three-dimensional model of the conformational state of the protein. The method is illustrated by refining the structure of the voltage-sensing domain (VSD) of the Kv1.2 potassium channel in the resting state and by exploring the distance histograms between spin-labels attached to T4 lysozyme. The resulting VSD structures are in good agreement with the consensus model of the resting state VSD and the spin-spin distance histograms from ESR/DEER experiments on T4 lysozyme are accurately reproduced.

How to fully apply the characteristics and advantages of light in numerical computation is an important issue that attracts many scholars. Though much research has been done in this field, how to design and realize specific applications or devices is still an issue to be solved. Based on this, we present an architecture and implementation method of modified signed-digit (MSD) optical adder from the point of applicability. In the implementation, we fully consider the different procedures of the MSD addition which including optical logical operation, results decoding, special storage area design, data feedback, control of light path, etc. Meanwhile, we also introduce pipeline mechanism which guarantees that the addition operation is an automatic and continuous process. This is a carry free adder design method which guarantees the addition has high data throughput. It is very suitable to fulfill the large-scale numerical computation. The experiment shows, the MSD adder not only has a reasonable and correct design, but also has high throughput rate, can work efficiently and steadily.