Searcharxiv⌕ Search

arXiv subjects

Jun Li

Publications and source records attributed to Jun Li.

At least 793 records · Page 44Linked to original sources
arXiv2019

Deep learning-based phase control method for coherent beam combining and its application in generating orbital angular momentum beams

We incorporate deep learning (DL) into coherent beam combining (CBC) systems for the first time, to the best of our knowledge. Using a well-trained convolutional neural network DL model, the phase error in CBC systems could be accurately estimated and preliminarily compensated. Then, the residual phase error is further compensated by stochastic parallel gradient descent (SPGD) algorithms. The two-stage phase control strategy combined with DL and SPGD algorithms is validated to be a feasible and promising technique to alleviate the long-standing problem that the phase control bandwidth decreases as the number of array elements expands. Further investigation denotes that the proposed phase control technique could be employed to generate orbital angular momentum (OAM) beams with different orders by distinguishing the OAM beams of conjugated phase distributions.

physics.optics↗
arXiv2019

Scattering Mechanisms and Modeling for Terahertz Wireless Communications

This paper provides an analysis of radio wave scattering for frequencies ranging from the microwave to the Terahertz band (e.g., 1 GHz - 1 THz), by studying the scattering power reradiated from various types of materials with different surface roughnesses. First, fundamentals of scattering and reflection are developed and explained for use in wireless mobile radio, and the effect of scattering on the reflection coefficient for rough surfaces is investigated. Received power is derived using two popular scattering models - the directive scattering (DS) model and the radar cross section (RCS) model through simulations over a wide range of frequencies, materials, and orientations for the two models, and measurements confirm the accuracy of the DS model at 140 GHz. This paper shows that scattering can become a prominent propagation mechanism as frequencies extend to millimeter-wave (mmWave) and beyond, but at other times can be treated like simple reflection. Knowledge of scattering effects is critical for appropriate and realistic channel models, which further support the development of massive multiple input-multiple output (MIMO) techniques, localization, ray tracing tool design, and imaging for future 5G and 6G wireless systems.

eess.SP↗
arXiv2019

BCOV's Feynman rule of quintic $3$-folds

We prove the BCOV Feynman rule by identifying the Feynman graph sum to the graph sum of an R-matrix action extracted from the NMSP theory. As direct consequences, (i) we obtain the genus one and genus two potentials, and (ii) we prove the two Yamaguchi-Yau equations.

math.AG↗
arXiv2019

Deep Discriminative Representation Learning with Attention Map for Scene Classification

Learning powerful discriminative features for remote sensing image scene classification is a challenging computer vision problem. In the past, most classification approaches were based on handcrafted features. However, most recent approaches to remote sensing scene classification are based on Convolutional Neural Networks (CNNs). The de facto practice when learning these CNN models is only to use original RGB patches as input with training performed on large amounts of labeled data (ImageNet). In this paper, we show class activation map (CAM) encoded CNN models, codenamed DDRL-AM, trained using original RGB patches and attention map based class information provide complementary information to the standard RGB deep models. To the best of our knowledge, we are the first to investigate attention information encoded CNNs. Additionally, to enhance the discriminability, we further employ a recently developed object function called "center loss," which has proved to be very useful in face recognition. Finally, our framework provides attention guidance to the model in an end-to-end fashion. Extensive experiments on two benchmark datasets show that our approach matches or exceeds the performance of other methods.

cs.CV↗
arXiv2019

Dust Extinction towards the Type Ia Supernova 2012cu in NGC 4772

Using photometric and spectroscopic data of Supernova (SN) 2012cu, a fairly reddened type Ia supernova, we derived its color excess curves and probed the dust extinction in its host galaxy, NGC 4772. In order to derive the extinction as a function of wavelength (i.e., $A_λ$), we model the color excess curves of SN 2012cu in terms of dust models consisting of silicate and carbonaceous (graphite or amorphous carbon) dust. The modeled extinction law towards SN 2012cu extends flatly to the far-ultraviolet (UV) bands, which is much flatter than those of the Milky Way and Magellanic Clouds, and the 2175 $Å$ feature is very weak or absent. The flatness of the modeled extinction curve in the UV bands suggests a "grey" extinction law of the active galactic nucleus in the vicinity of the SN 2012cu-Earth line of sight. Our results indicate that the sizes of the dust in the ISM towards SN 2012cu in NGC 4772 are larger than those of the Milky Way and the Large Magellanic Cloud, and much larger than that of the Small Magellanic Cloud. The best fitting gives an observed visual extinction towards SN 2012cu of $A_V \approx 2.6$ mag, a reddening of $E(B-V) \approx 1.0$ mag, with a total-to-selective extinction ratio $R_V \approx 2.7$, consistent with previous results.

astro-ph.GA↗
arXiv2019

Local-measurement-based quantum state tomography via neural networks

Quantum state tomography is a daunting challenge of experimental quantum computing even in moderate system size. One way to boost the efficiency of state tomography is via local measurements on reduced density matrices, but the reconstruction of the full state thereafter is hard. Here, we present a machine learning method to recover the full quantum state from its local information, where a fully-connected neural network is built to fulfill the task with up to seven qubits. In particular, we test the neural network model with a practical dataset, that in a 4-qubit nuclear magnetic resonance system our method yields global states via the 2-local information with high accuracy. Our work paves the way towards scalable state tomography in large quantum systems.

quant-ph↗
arXiv2018

Rotation invariant singular Kähler metrics with constant scalar curvature on $\mathbb{C}^n$

The scalar curvature equation for rotation invariant Kähler metrics on $\mathbb{C}^n \backslash \{0\}$ is reduced to a system of ODEs of order 2. By solving the ODEs, we obtain complete lists of rotation invariant zero or positive csck on $\mathbb{C}^n \backslash \{0\}$ in lower dimensions. We also prove that there does not exist negative csck on $\mathbb{C}^n \backslash \{0\}$ for $n=2,3$.

math.DG↗
arXiv2018

Experimental Implementation of Efficient Quantum Pseudorandomness on a 12-spin System

Quantum pseudorandomness, also known as unitary designs, comprise a powerful resource for quantum computation and quantum engineering. While it is known in theory that pseudorandom unitary operators can be constructed efficiently, realizing these objects in realistic physical systems can be a challenging task. In this work, we study quantum pseudorandomness generation on a 12-spin nuclear magnetic resonance system. The experimental process is based on the recently proposed design Hamiltonian approach, which has the merit of being significantly more efficient than previous protocols. By applying random refocusing sequences to the experimental system we create a design Hamiltonian the dynamics of which quickly forms unitary designs. We then use multiple-quantum techniques to measure spreading of quantum coherences over system's degrees of freedom, and so to probe the growth of quantum pseudorandomness. The measured multiple-quantum coherence spectra indicate that substantial quantum pseudorandomness have been achieved.

quant-ph↗
arXiv2018

Covert Wireless Communications with Channel Inversion Power Control in Rayleigh Fading

In this work, we adopt channel inversion power control (CIPC) to achieve covert communications aided by a full-duplex receiver. Specifically, the transmitter varies the power and phase of transmitted signals as per the channel to the receiver, such that the receiver can decode these signals without knowing the channel state information. This eliminates the required feedback from the transmitter to the receiver, which aids hiding the transmitter from a warden. The truncated CIPC and conventional CIPC schemes are proposed and examined, where for truncated CIPC covert transmission ceases when the channel quality from the transmitter to the receiver is low, while for conventional CIPC covert transmission always occurs regardless of this channel quality. We examine their performance in terms of the achieved effective covert throughput (ECT), which quantifies the amount of information that the transmitter can reliably convey to the receiver, subject to the constraint that the warden's detection error probability is no less than some specific value. Our examination shows that the truncated CIPC scheme can outperform the conventional CIPC scheme due to this constraint.

cs.IT↗
arXiv2018

Machine-Learning-based High-resolution DOA Measurement and Robust DM for Hybrid Analog-Digital Massive MIMO Transceiver

At hybrid analog-digital (HAD) transceiver, an improved HAD rotational invariance techniques (ESPRIT), called I-HAD-ESPRIT, is proposed to measure the direction of arrival (DOA) of desired user, where the phase ambiguity due to HAD structure is addressed successfully. Subsequently, a machine-learning (ML) framework is proposed to improve the precision of measuring DOA. In the training stage, the HAD transceiver works as a receiver and repeatedly measures the values of DOA via I-HAD-ESPRIT to form a slightly large training data set (TDS) of DOA. From TDS, we find that the probability density function (PDF) of DOA measurement error (DOAME) is approximated as a Gaussian distribution by the histogram method in ML. This TDS is used to learn the mean of DOA and the variance of DOAME, which are utilized to infer their values and improve their precisions in the real-time stage. The HAD transceiver rapidly measures the real-time value of DOA some times to form a relatively small real-time set (RTS), which is used to learn the real-time mean and variance of DOA/ DOAME. Then, three weight combiners are proposed to combine the-maximum-likelihood-learning outputs of TDS and RTS. Their weight factors depend intimately on the numbers of elements in TDS and RTS, and signal-to-noise ratios during the two stages. Using the mean and variance of DOA/DOAME, their PDFs can be given directly, and we propose a robust beamformer for directional modulation (DM) transmitter with HAD by fully exploiting the PDF of DOA/DOAME, especially a robust analog beamformer on RF chain. Simulation results show that: 1) The proposed I-HAD-ESPRIT can achieve the HAD CRLB; 2) The proposed ML framework performs much better than the corresponding real-time one without training stage, 3) The proposed robust DM transmitter can perform better than the corresponding non-robust ones in terms of bit error rate and secrecy rate.

cs.IT↗
arXiv2018

Accurate Spectral Super-resolution from Single RGB Image Using Multi-scale CNN

Different from traditional hyperspectral super-resolution approaches that focus on improving the spatial resolution, spectral super-resolution aims at producing a high-resolution hyperspectral image from the RGB observation with super-resolution in spectral domain. However, it is challenging to accurately reconstruct a high-dimensional continuous spectrum from three discrete intensity values at each pixel, since too much information is lost during the procedure where the latent hyperspectral image is downsampled (e.g., with x10 scaling factor) in spectral domain to produce an RGB observation. To address this problem, we present a multi-scale deep convolutional neural network (CNN) to explicitly map the input RGB image into a hyperspectral image. Through symmetrically downsampling and upsampling the intermediate feature maps in a cascading paradigm, the local and non-local image information can be jointly encoded for spectral representation, ultimately improving the spectral reconstruction accuracy. Extensive experiments on a large hyperspectral dataset demonstrate the effectiveness of the proposed method.

cs.CV↗
arXiv2018

Quantum singularity theory via cosection localization

We generalize the cosection localized Gysin map to intersection homology and Borel-Moore homology, which provides us with a purely topological construction of the Fan-Jarvis-Ruan-Witten invariants and some GLSM invariants.

math.AG↗
arXiv2018

Multi-step Steady-State Measurements of Low Permeability Using Series Circuit with A Reference Rock Sample

A multi-step steady-state (MSSS) method is proposed here for the measurement of low permeability. This new method can accurately and easily measure very low permeabilities of rock samples using a new setup, where the targeted rock sample and ordinary apparatus components are connected with a reference rock sample to form a series circuit. Any conventional rock sample with high permeability could be used as a reference rock sample such that the traditional steady-state measurement is feasible to accurately determine its permeability as a reference value in the MSSS method. The challenging measurement of tiny mass flux rate by advanced pump system is avoided and the permeability of targeted rock sample can be directly computed using the pressure drops, sectional areas and lengths of the two connected rock samples, and the known permeability of the reference rock sample, based on the mass conservation principle in a series circuit at steady state. Multi-step measurements using additional reference rock samples will be needed if the pressure drop across the first reference rock sample is too small to be accurately measured due to high permeability ratio when it is connected with the targeted rock sample to form a series circuit. The relative pressure drops can be small since the measurement of flow speed is unnecessary, which improves the accuracy in studying the dependence of gas permeability on the pore pressure. Consequently, the advantages of the MSSS method include low expense, simplicity, high accuracy and efficiency.

physics.geo-ph↗
arXiv2018

Quasi-incompressible Multi-species Ionic Fluid Models

In traditional hydrodynamic theories for ionic fluids, conservation of the mass and linear momentum is not properly taken care of. In this paper, we develop hydrodynamic theories for a viscous, ionic fluid of $N$ ionic species enforcing mass and momentum conservation as well as considering the size effect of the ionic particles. The theories developed are quasi-incompressible in that the mass-average velocity is no longer divergence-free whenever there exists variability in densities of the fluid components, and the models are dissipative. We present several ways to derive the transport equations for the ions, which lead to different rates of energy dissipation. The theories can be formulated in either number densities, volume fractions or mass densities of the ionic fluid components. We show that the theory with the Cahn-Hilliard transport equation for ionic species reduces to the classical Poisson-Nernst-Planck (PNP) model with the size effect for ionic fluids when the densities of the fluid components are equal and the entropy of the solvent is neglected. It further reduces to the PNP model when the size effect is neglected. A linear stability analysis of the model together with two of its limits, which is the extended PNP model (EPNP defined in the text) and the classical PNP model (CPNP) with the finite size effect, on a constant state and a comparison among the three models in 1D space are presented to highlight the similarity and the departure of this model from the EPNP and the CPNP model.

physics.flu-dyn↗
arXiv2018

Achieving Covert Wireless Communications Using a Full-Duplex Receiver

Covert communications hide the transmission of a message from a watchful adversary while ensuring a certain decoding performance at the receiver. In this work, a wireless communication system under fading channels is considered where covertness is achieved by using a full-duplex (FD) receiver. More precisely, the receiver of covert information generates artificial noise with a varying power causing uncertainty at the adversary, Willie, regarding the statistics of the received signals. Given that Willie's optimal detector is a threshold test on the received power, we derive a closed-form expression for the optimal detection performance of Willie averaged over the fading channel realizations. Furthermore, we provide guidelines for the optimal choice of artificial noise power range, and the optimal transmission probability of covert information to maximize the detection errors at Willie. Our analysis shows that the transmission of artificial noise, although causes self-interference, provides the opportunity of achieving covertness but its transmit power levels need to be managed carefully. We also demonstrate that the prior transmission probability of 0.5 is not always the best choice for achieving the maximum possible covertness, when the covert transmission probability and artificial noise power can be jointly optimized.

cs.IT↗
arXiv2018

Second Order Linear Energy Stable Schemes for Allen-Cahn Equations with Nonlocal Constraints

We present a set of linear, second order, unconditionally energy stable schemes for the Allen-Cahn equation with nonlocal constraints that preserves the total volume of each phase in a binary material system. The energy quadratization strategy is employed to derive the energy stable semi-discrete numerical algorithms in time. Solvability conditions are then established for the linear systems resulting from the semi-discrete, linear schemes. The fully discrete schemes are obtained afterwards by applying second order finite difference methods on cell-centered grids in space. The performance of the schemes are assessed against two benchmark numerical examples, in which dynamics obtained using the volumepreserving Allen-Cahn equations with nonlocal constraints is compared with those obtained using the classical Allen-Cahn as well as the Cahn-Hilliard model, respectively, demonstrating slower dynamics when volume constraints are imposed as well as their usefulness as alternatives to the Cahn-Hilliard equation in describing phase evolutionary dynamics for immiscible material systems while preserving the phase volumes. Some performance enhancing, practical implementation methods for the linear energy stable schemes are discussed in the end.

math.NA↗