Cognitive radio is a promising technology to improve spectral efficiency. However, the secure performance of secondary network achieved by using physical layer security techniques limited its transmit power and channel fading. In order tackle this issue, cognitive unmanned aerial vehicle (UAV) communication studied exploiting high flexibility UAV possibility establishing line-of-sight links. The average secrecy rate maximized robustly optimizing UAV's trajectory power. Our problem...

Wireless powered techniques have been recognized as promising in future wireless communication systems, especially cognitive radios (CRs) with energy-limit devices. However, most of the existing works focus on CRs an ideal linear energy harvesting model. In this paper, a wideband CR network is considered, and practical non-linear model adopted. To maximize sum throughput secondary users, time, channel allocation, transmit power are jointly optimized. The closed-form expressions for optimal...

The integrated radar and communication system is promising in the next generation wireless networks. However, its performance confined by limited energy. In order to overcome this, a powered proposed. An energy minimization problem formulated subject constraints on performances. beamforming radar-communication waveform are jointly optimized minimize consumption challenging non-convex solved using semidefinite relaxation auxiliary variable methods. It proved that optimal solution can be...

This paper proposes a generic method to learn interpretable convolutional filters in deep neural network (CNN) for object classification, where each filter encodes features of specific part. Our does not require additional annotations parts or textures supervision. Instead, we use the same training data as traditional CNNs. automatically assigns high conv-layer with an part certain category during learning process. Such explicit knowledge representations conv-layers CNN help people clarify...

Given the obstacle in accentuating reconstruction accuracy for diagnostically significant tissues, most existing MRI methods perform targeted of entire MR image without considering fine details, especially when dealing with highly under-sampled images. Therefore, a considerable volume efforts has been directed towards surmounting this challenge, as evidenced by emergence numerous dedicated to preserving high-frequency content well textural details reconstructed image. In case, exploring...

This paper presents a novel method for diagnosing faults using fault tree analysis and Bayesian networks (BN) to optimize system diagnosis. All minimal cut sets were generated via qualitative of an efficient zero-suppressed binary decision diagram (ZBDD), while the diagnostic importance factor (DIF) components calculated by mapping into equivalent BN. Also, these results such as DIF updated after receiving evidence data from sensors used develop algorithm. Furthermore, (DDT) was guide...

This paper proposes a method to disentangle and quantify interactions among words that are encoded inside DNN for natural language processing. We construct tree encode salient extracted by the DNN. Six metrics proposed analyze properties of between constituents in sentence. The interaction is defined based on Shapley values words, which considered as an unbiased estimation word contributions network prediction. Our used BERT, ELMo, LSTM, CNN, Transformer networks. Experimental results have...

Numerous learning-based methods have been proposed to address the challenging task of solving inverse scattering problems (ISPs), with requirement ample labeled training datasets. To overcome difficulty collecting fully datasets, we present a physically guided deep unfolding parallel network self-supervised learning strategy solve nonlinear ISPs, called SSISP-Net. The method incorporates physical model and iterative optimization algorithm ISPs into architecture and, thus, forms network....

In this article, to bridge the gap between traditional model-based methods and data-driven deep learning schemes, we propose a physical model-inspired unrolling network for solving nonlinear inverse scattering problems, termed PM-Net. The proposed end-to-end is formed by two consequent steps. First, an augmented Lagrangian method introduced transform constrained objective function be unconstrained optimization. addition, it further decomposed into four quasi-linear subproblems. Second,...

Electromagnetic inverse scattering problems (ISPs) aim to retrieve permittivities of dielectric scatterers from the measurement. It is often highly nonlinear, caus-ing problem be very difficult solve. To alleviate issue, this letter exploits a linear model-based network (LMN) learning strategy, which benefits both model complexity and data learning. By introducing for ISPs, new with network-driven regular-izer proposed. For attaining efficient end-to-end learning, architecture...

The clutter encountered in the ground-penetrating radar (GPR) system severely obscures visibility of subsurface objects, especially case overlapping target responses and clutter. In this letter, a novel self-supervised learning strategy with dual-network architecture pseudolabels is proposed. First, consists two subnetworks: one to simulate low-rank part, another simulates sparse part. Second, raw GPR data are decomposed as sum matrices by robust nonnegative matrix factorization (RNMF),...

Lensless imaging shifts the burden of from bulky and expensive hardware to computing, which enables new architectures for portable cameras. However, twin image effect caused by missing phase information in light wave is a key factor limiting quality lensless imaging. Conventional single-phase encoding methods independent reconstruction separate channels pose challenges removing images preserving color fidelity reconstructed image. In order achieve high-quality imaging, multiphase via...

A CPW-fed ultra-wideband antenna was designed. The etched on a single-layer copper-cladding substrate, of which the material FR4 with relative permittivity 4.4, and magnitude 40.0 mm × 50.0 1.6 mm. parameters are simulated optimized HFSS. This paper proposes new trapezoidal UWB that bandwidth (return loss ≤ ?10 dB) covers 2.7 - 9.3 GHz range, means 110% good radiation patterns gain. Simulated measured results for return loss, pattern gain were presented. agreement has been obtained between...

Seismic full-waveform inversion (FWI) is a powerful technique for high-resolution imaging of subsurface physical properties. However, it suffers from the possibility falling into local minimum due to inherent nonlinearity and ill-posedness. Recently, data-driven deep learning approach has been used solve ill-posed inverse problems, with limitations high data collection costs poor model generalization capabilities. To alleviate these difficulties, we unroll iterative optimization algorithm...

Targets location and parameter inversion are always active research field of Ground Penetrating Radar (GPR) useful to address some challenges in civil military applications. Since the amplitude delay receiving signal could correspondingly change due varying dimension, material targets, permittivity background. So, this paper, we present a new framework integrated Generalized Hough Transform (GHT) with neural network reconstruct their non-linear relationship implement targets inversion. The...

Although metal halide perovskites (MHPs) have demonstrated remarkable external quantum efficiencies (EQEs) in red and green light-emitting diodes (LEDs), the blue ones confront efficiency stability problems due to high defect density perovskite films. Large amounts of passivation strategies are successfully developed improve device performance. Nevertheless, influence molecular configuration passivators on crystallization process has not been comprehensively investigated so far. Here, we...

Single photon counting compressive imaging, a combination of single-pixel-imaging and single-photon-counting technology, is provided with low cost ultra-high sensitivity. However, it requires long imaging time when applying traditional compressed sensing (CS) reconstruction algorithms. A deep-learning-based network refrains iterative computation while achieving efficient reconstruction. This paper proposes (OGTM) based on generative model, adding sampling sub-network to achieve...

Compressive sensing theory have been proposed in the field of radar for target detection. The challenge compressive applied to passive bistatic lies high computational complexity aggravated by transmitted signal dependent time-varying sparse model. So, we propose a transmitted-signal-free and time-invariant model based on orthogonal frequency division multiplexing waveforms. We first generate using surveillance pilot information only, exploiting sparsity scene including only few targets...

Determination of background permittivity and removing the direct wave is one main problems in data processing on Ground Penetrating Radar. The interference will greatly affect accuracy resolution GPR imaging. We propose an f-k filtering approach for solution, which use a difference apparent velocity between reflection from target. Besides, Stolt migration algorithm combined F-K filter with minimum entropy technique proposed to improve focus effect radar image realize precise inversion...