This paper deals with the joint design problem of transmit waveform and receive filter for robust detection ground moving target multiple-input multiple-output Space time adaptive processing (MIMO- STAP) radar in presence clutter uncertainties. With prior knowledge statistics, averaged signal-to-interference-plus-noise ratio (SINR) is formulated as a figure merit to maximize. The problems continuous discrete phase cases, respectively, are constant-modulus similarity constraints. Then, an...

Mode recognition is a basic task to interpret the behavior of multi-functional radar. The existing methods need train complex and huge neural networks improve ability, it difficult deal with mismatch between training set test set. In this paper, learning framework based on residual network (ResNet) support vector machine (SVM) designed, solve problem mode for non-specific radar, called multi-source joint (MSJR). key idea embed prior knowledge radar into model, combine manual intervention...

Centella asiatica (L.) Urban is a tropical medicinal plant with long history of therapeutic uses. Madecassic acid and terminolic acid, pair structural isomers, are two constituents asiatica. A method using reversed-phase high performance liquid chromatography in which β-cyclodextrin (β-CD) was the additive mobile phase has been developed for separation determination isomers. The compounds can be isolated resolution on C18 column addition β-CD phase. mechanism isomers discussed. It assumed...

Abstract Radar signal recognition plays a vital role in electronic warfare. For the multifunction radars (MFRs) with complex dynamical modes, needs to identify not only emitter but also its current functional state. Existing research on MFR mainly focuses hierarchical modelling approaches. Inspired by recent progress of deep neural networks, authors propose further develop radar recurrent networks. Here, more efficient method for state MFRs based gated unit (GRU). The makes full use ability...

Radar emitter signal recognition under noisy background is one of the focus areas in research on radar processing. In this study, soft thresholding function embedded into deep learning network models as a novel nonlinear activation function, achieving advanced results. Specifically, an sub-network used to learn threshold according input feature, which results each feature having its own independent function. Compared with conventional functions, characterized by flexible conversion and...

With the development of multifunction radar, traditional specific emitter identification (SEI) can no longer meet needs observe-orient-decide-act (OODA) closed loop, and most networks are in process converting one-dimensional (1D) radar signals into two-dimensional (2D) to adapt network input, which easily misses information. To address above problems, this paper adopts a 1D convolutional residual neural with block attention module (1D-CBAM-ResNet) for automatic learning single-step...

An iterative detection algorithm, based on fixed length sliding window, is proposed for multi-function phased array radar (MPAR) behavior recognition. First, characteristic parameters such as frequency, pulse width, amplitude, repetition interval and beam orientation were extracted. Then the sequence divided by appropriate fixed-length window. Finally, conditional probability calculated step Bayesian criterion, result compared with previous to determine whether it a change point. This...

In future electronic warfare (EW), there will be many unmanned aerial vehicles (UAVs) equipped with support measure (ESM) systems, which often encounter the challenge of radar emitter identification (REI) few labeled samples. To address this issue, we propose a novel deep learning network, IRelNet, could easily embedded in computer system UAV. This network was designed channel attention, spatial attention and skip-connect features, meta-learning technology applied to solve REI problem....

To increase the number of estimable signal sources, two-parallel nested arrays are proposed, which consist two subarrays with M sensors, and can estimate two-dimensional (2-D) direction arrival (DOA) 2 sources. solve problem finding arrays, a 2-D DOA estimation algorithm based on sparse Bayesian is proposed. Through vectorization matrix, smoothing reconstruction matrix singular value decomposition (SVD), reduces size dictionary data noise. A learning used to one dimension angle. By joint...

Signal features can be obscured in noisy environments, resulting low accuracy of radar emitter signal recognition based on traditional methods. To improve the ability learning from signals, a new method one-dimensional (1D) deep residual shrinkage network (DRSN) is proposed, which offers following advantages: (i) Unimportant are eliminated using soft thresholding function, and thresholds automatically set attention mechanism; (ii) without any professional knowledge processing or dimension...

Target recognition mainly focuses on three approaches: optical-image-based, echo-detection-based, and passive signal-analysis-based methods. Among them, the signal-based method is closely integrated with practical applications due to its strong environmental adaptability. Based radar signal analysis, we design an “end-to-end” model that cascades a noise estimation network identify working modes in noisy environments. The implemented based U-Net, which adopts of feature extraction...

  The precise point positioning (PPP) timing service has been proposed in recent years and demonstrated to achieve sub-nanosecond accuracy. As an integral part of BDS-3, the PPP services via b2b signals (hereafter referred as PPP-B2b) are provided by BDS-3 GEO satellites with better than decimeter-level accuracy for users around China. In this research, a high-precision receiver based on PPP-B2b established, which obtains local clock offset respect Beidou Time (BDT) through time...

This study aims to advance medical research by identifying and segmenting functional tissue units (FTUs) within five human organs using deep learning techniques. The dataset comprises slice images from the Human Protein Atlas (HPA) BioMolecular Program (HuBMAP). We assess segmentation accuracy mean Dice coefficient. Analysis indicates significant distributional differences across gender age groups, prompting design of varied sample weighting coefficients sampling embedding strategies....

Reconnaissance unmanned aerial vehicles are specifically designed to estimate parameters and process intercepted signals for the purpose of identifying locating radars. However, distinguishing quasi-simultaneous arrival (QSAS) has become increasingly challenging in complex electromagnetic environments. In order address problem, a framework self-supervised deep representation learning is proposed. The consists two phases: (1) pre-train an autoencoder. For unlabeled QSAS representation,...

The radar signal intrapulse clustering (RSIPC) can help achieve unsupervised emitter identification, which is of great significance in the field electronic warfare. In order to address poor performance traditional methods handling RSIPC tasks, we propose a contrastive learning-based method called CLIPC. Since single-domain information intrapulses may result loss important features, integrate multidomain obtain fusion samples. By training learning network on these samples, extract deep...

We innovatively apply three different methods of encoding signal as 2-D plots to radar emitter recognition: Recurrence Plots (RP), Gramian Angular Field (GAF) and Markov Transition (MTF), thus the recognition problem is converted into image processing problem. build a 2-stage convolutional neural network (CNN) model make use its mature technology in field computer vision for recognition. These pipeline offers following advantages: i) Encoding enable us visualize certain aspects signals...

With the widespread use of multifunction radars (MFRs), it is hard for traditional radar signal recognition technology to meet needs current electronic intelligence systems. For an MFR, necessary identify not only type or individual emitter but also its state. Existing methods MFR states through hierarchical modeling, most them rely heavily on prior information. In paper, we focus state with actual intercepted signals and develop by introducing recurrent neural networks (RNNs) deep learning...