A climate/vegetation model simulates episodic wetter and drier periods at the 21,000-y precession period in eastern North Africa, Arabian Peninsula, Levant over past 140,000 y. Large orbitally forced wet/dry extremes occur during interglacial time, ∼130 to 80 ka, conditions between these two prevail glacial ∼70 15 ka. Orbital causes high seasonality Northern Hemisphere (NH) insolation ∼125, 105, 83 with stronger northward extended summer monsoon rains Africa Peninsula increased winter...

The frequency diverse array (FDA) radar has drawn great attention due to the periodicity of beampattern in range, angle, and time. In this letter, we restudy recent work that designed a time-invariant FDA radar, which can focus transmit energy desired position. We reanalyze derivation synthesis point out neglected constraint condition, leads research an impractical direction. By comparing replication one prior works with our result, draw conclusion it is impossible obtain merely focusing on...

Due to the influence of complex marine environment, target detection based on statistical theory is difficult achieve high-performance. Moreover, due various targets' motion characteristics, only using a single feature for unreliable. In this paper, from perspective extraction and classification, sea clutter are classified by deep learning methods. To required false alarm rate, dual-channel convolutional neural networks (DCCNN) false-alarm-controllable classifier (FACC)-based method...

Attention has been focused on the moving target detection in heavy sea clutter. On basis of model with fluctuant amplitudes, a novel adaptive algorithm fractional Fourier transform (FRFT) domain is proposed, which combines statistic-based and FRFT-based method. FRFT good energy concentration property linear frequency modulation (LFM) signal optimal angle, determined by calculating spectral kurtosis (SK) domain. Grading iterative search method used for accuracy parameter estimation fast...

The radar echoes of moving targets in a clutter background are extremely weak and their characteristics complex. A low-observable target detection technology has become key constraint to performance, requiring the provide flexible freedom higher parameter estimation capabilities. In this paper, frequency diverse array MIMO (FDA-MIMO) is used for background, which combines information spatial (azimuth)range-frequency (Doppler) domains. Based on signal model FDA-MIMO radar, novel processing...

Introduction of a dialkoxy-substituted tetraphenylethylene entity into porphyrin sensitizers improves the photovoltages and efficiencies dye sensitized solar cells, affording high power conversion efficiency 12.3%.

Due to the complex sea clutter environment and target features, conventional statistical theory-based methods cannot achieve high performance in maritime detection tasks. Conventional deep learning, such as convolutional neural networks (CNNs)-based process each signal sample independently, temporal-spatial domain correlation information is seldom used. To full utilization of contained radar signals improve performance, a graph network (GCN) considered, which has shown great advantages data...

Effective detection of low observable moving target at sea is important for remote sensing and radar signal processing. The non‐Gaussian property clutter lack accurate model make the difficult statistics based detectors. Also fractal techniques in time domain cannot achieve high probability heavy clutter. To help solve problems, characteristics IPIX datasets fractional Fourier transform (FRFT) are analysed making use fluctuation FRFT amplitudes algorithms proposed on domain. Firstly,...

Radar maneuvering target detection in clutter background should not only consider the complex characteristics of to accumulate its energy as much possible, but also suppress improve signal-to-clutter ratio (SCR). The traditional fractional domain transform-based method requires parameters match searching, which costs heavy computational burden case a large amount data. Sparse FT and sparse can obtain high-resolution representation target, signal sparsity needs be known before, performance is...

In this letter, a novel fast detection algorithm, known as robust sparse fractional Fourier transform (RSFRFT), is proposed for low-observable maneuvering target in clutter background. The discrete FRFT (DFRFT)-based method time-consuming large data volumes and the performance of (SFRFT)-based algorithm will be significantly degraded heavy Using two levels detection, defects DFRFT SFRFT algorithms are overcome using algorithm. first-level performed on subsampled spectrum to estimate...

Abstract Three transient National Center for Atmospheric Research Community Climate System Model, version 3 model simulations were analyzed to study the responses of El Niño–Southern Oscillation (ENSO) and equatorial Pacific annual cycle (AC) external forcings over last 300,000 years. The time‐varying boundary conditions insolation, greenhouse gases, continental ice sheets, accelerated by a factor 100, sequentially added in these simulations. simulated ENSO AC amplitudes change phase, both...

This study mainly describes the fractal property of frequency spectrum sea clutter and application obtained characteristic in domain to constant-false-alarm-rate (CFAR) target detection within clutter. First, this takes fractional Brownian motion (FBM) for example, FBM is proved be theoretically on condition that time series fractal. argument lays foundation theory spectrum. Next, X- S-band real radar data are used verification Finally, effects length fast Fourier transform (FFT) analysed...

Marine target detection is always an important issue in the field of detection, and scholars at home abroad have achieved fruitful research success, but there are still some problems while using traditional method such as Large interference sea clutter on limited performance. Recently, deep learning has been rapidly developed widely applied However, currently no related application to navigation radar. In this paper, we propose a marine based improved Faster R-CNN for radar PPI (Plane...

As an outlook of the terrestrial network, flexible Unmanned Aerial Vehicle (UAV)-aided satellite-terrestrial integration network would have numerous prospective applications such as service enhancement, maritime communication, military, and emergency which is getting significant attention. However, public open-access must result in various imperative risks including impersonation, sensitive data, privacy disclosure. Due to several unique characteristics long transmission distance, unstable...

The research on maritime target detection signals has significant value in various fields. Conventional statistical theory-based methods are limited by the complex sea clutter environment and characteristics, making it challenging to achieve high-performance detection. In practical scenarios such as observation, radar observation area is expansive. And beam cannot remain fixed one direction for prolonged intervals. Consequently, not feasible accumulate multiple pulses within a single azimuth...

A technique to estimate tropical cyclone (TC) current intensity based on geostationary satellite infrared window (IRW) and water vapor (WV) imagery is explored in this paper. First, combine the advantages of IRW WV minus (WV-IRW) imagery, a WV-IRW-to-IRW ratio (WIRa)-based indicator proposed. This not only can display inner-core convection's symmetrization level vigor but also able screen out thin cirrus, stratospheric anomaly, overshooting tops from average deep convection. It highly...

Existing cold-start recommendation methods often adopt item-level alignment strategies to align the content feature and collaborative of warm items for model training, however, cold in test stage have no historical interactions with users obtain feature. These existing models ignore aforementioned condition training stage, resulting performance limitation. In this paper, we propose a preference aware dual contrastive learning based (PAD-CLRec), where user is explored take into account...

Traditional detection and tracking methods struggle with the complex dynamic maritime environment due to their poor generalization capabilities. To address this, this paper improves YOLOv5 network by integrating Transformer a Convolutional Block Attention Module (CBAM) multi-frame image information obtained from radar scans. It proposes method based on Detection Tracking Network (DTNet), which leverages transfer learning DeepSORT algorithm, enhancing capabilities of model across various...

Traditional analysis of the wireless network operation stability (WNOS) is based on several indicators or human experience, there a great difficulty in carrying out systematic evaluation. With rapid development scale and improving requirement fine management, an urgent need to build evaluation method WNOS. By using idea big data analysis, collecting various types from existing network, comprehensive solution built help current systematically periodically. This can further implementation...

Low‐observable manoeuvring target detection is a challenging problem for radar signal processing, due to the complex environment, manoeuvrability, limited observation time etc. The study proposed novel long‐time coherent integration (LTCI) method target, especially low‐observable unmanned aerial vehicle targets. fast non‐parametric searching LTCI via non‐uniform (NU) resampling and scale processing (SP) technique proposed, where high‐order phase reduced linear term using NU resampling. After...