arXiv subjects

Jian Zhao

Publications and source records attributed to Jian Zhao.

324 records · Page 18Linked to original sources

arXiv2017

Generalized-impedance and Stability Criterion for Grid-connected Converters

The output impedance matrix of a grid-connected converter plays an important role in analyzing system stability. Due to the dynamics of the DC-link control and the phase locked loop (PLL), the output impedance matrices of the converter and grid are difficult to be diagonally decoupled simultaneously, neither in the dq domain nor in the phase domain. It weakens the effectiveness of impedance-based stability criterion (ISC) in system oscillation analysis. To this end, this paper innovatively proposes the generalized-impedance based stability criterion (GISC) to reduce the dimension of the transfer function matrix and simplify system small-signal stability analysis. Firstly, the impedances of the converter and the grid in polar coordinates are formulated, and the concept of generalized-impedance of the converter and the grid is put forward. Secondly, through strict mathematical derivation, the equation that implies the dynamic interaction between the converter and the grid is then extracted from the characteristic equation of the grid-connected converter system. Using the proposed method, the small-signal instability of system can be interpreted as the resonance of the generalized-impedances of the converter and the grid. Besides, the GISC is equivalent to ISC when the dynamics of the outer-loop control and PLL are not considered. Finally, the effectiveness of the proposed method is further verified using the MATLAB based digital simulation and RT-LAB based hardware-in-the-loop (HIL) simulation.

eess.SY↗

arXiv2016

Robust LSTM-Autoencoders for Face De-Occlusion in the Wild

Face recognition techniques have been developed significantly in recent years. However, recognizing faces with partial occlusion is still challenging for existing face recognizers which is heavily desired in real-world applications concerning surveillance and security. Although much research effort has been devoted to developing face de-occlusion methods, most of them can only work well under constrained conditions, such as all the faces are from a pre-defined closed set. In this paper, we propose a robust LSTM-Autoencoders (RLA) model to effectively restore partially occluded faces even in the wild. The RLA model consists of two LSTM components, which aims at occlusion-robust face encoding and recurrent occlusion removal respectively. The first one, named multi-scale spatial LSTM encoder, reads facial patches of various scales sequentially to output a latent representation, and occlusion-robustness is achieved owing to the fact that the influence of occlusion is only upon some of the patches. Receiving the representation learned by the encoder, the LSTM decoder with a dual channel architecture reconstructs the overall face and detects occlusion simultaneously, and by feat of LSTM, the decoder breaks down the task of face de-occlusion into restoring the occluded part step by step. Moreover, to minimize identify information loss and guarantee face recognition accuracy over recovered faces, we introduce an identity-preserving adversarial training scheme to further improve RLA. Extensive experiments on both synthetic and real datasets of faces with occlusion clearly demonstrate the effectiveness of our proposed RLA in removing different types of facial occlusion at various locations. The proposed method also provides significantly larger performance gain than other de-occlusion methods in promoting recognition performance over partially-occluded faces.

cs.CV↗

arXiv2016

Runge-Kutta discontinuous Galerkin methods for the special relativistic magnetohydrodynamics

This paper develops $P^K$-based non-central and central Runge-Kutta discontinuous Galerkin (DG) methods with WENO limiter for the one- and two-dimensional special relativistic magnetohydrodynamical (RMHD) equations, $K=1,2,3$. The non-central DG methods are locally divergence-free, while the central DG are "exactly" divergence-free but have to find two approximate solutions defined on mutually dual meshes. The adaptive WENO limiter first identifies the "troubled" cells by using a modified TVB minmod function, and then uses the WENO technique to locally reconstruct a new polynomial of degree $(2K+1)$ inside the "troubled" cells replacing the DG solution by based on the cell average values of the DG solutions in the neighboring cells as well as the original cell averages of the "troubled" cells. The WENO limiting procedure does not destroy the locally or "exactly" divergence-free property of magnetic field and is only employed for finite "troubled" cells so that the computational cost can be as little as possible. Several test problems in one and two dimensions are solved by using our non-central and central Runge-Kutta DG methods with WENO limiter. The numerical results demonstrate that our methods are stable, accurate, and robust in resolving complex wave structures.

math.NA↗

arXiv2016

Augmenting Mobile Phone Interaction with Face-Engaged Gestures

The movement of a user's face, easily detected by a smartphone's front camera, is an underexploited input modality for mobile interactions. We introduce three sets of face-engaged interaction techniques for augmenting the traditional mobile inputs, which leverages the combination of the head movements with touch gestures and device motions, all sensed via the phone's built-in sensors. We systematically present the space of design considerations for mobile interactions using one or more of the three input modalities (i.e., touch, motion, and head). The additional affordances of the proposed techniques expand the mobile interaction vocabulary, and can facilitate unique usage scenarios such as one-hand or touch-free interaction. An initial evaluation was conducted and users had positive reactions to the new techniques, indicating the promise of an intuitive and convenient user experience.

cs.HC↗

arXiv2016

Central Runge-Kutta discontinuous Galerkin methods for the special relativistic hydrodynamics

This paper developes Runge-Kutta $P^K$-based central discontinuous Galerkin (CDG) methods with WENO limiter to the one- and two-dimensional special relativistic hydrodynamical (RHD) equations, $K=1,2,3$. Different from the non-central DG methods, the \CDG{} have to find two approximate solutions defined on mutually dual meshes. For each mesh, the CDG approximate solutions on its dual mesh are used to calculate the flux values in the cell and on the cell boundary so that the approximate solutions on mutually dual meshes are coupled with each other, and the use of numerical flux may be avoided. The WENO limiter is adaptively implemented via two steps: the "troubled" cells are first identified by using a modified TVB minmod function, and then the WENO technique is used to locally reconstruct new polynomials of degree $(2K+1)$ replacing the CDG solutions inside the "troubled' cells by the cell average values of the CDG solutions in the neighboring cells as well as the original cell averages of the "troubled" cells. Because the WENO limiter is only employed for finite "troubled" cells, the computational cost can be as little as possible. The accuracy of the CDG without the numerical dissipation is analyzed and calculation of the flux integrals over the cells is also addressed. Several test problems in one and two dimensions are solved by using our \CDG{} with WENO limiter. The computations demonstrate that our methods are stable, accurate, and robust in solving complex RHD problems.

math.NA↗

arXiv2016

Orthogonal Chirp Division Multiplexing

Chirp waveform plays a significant role in radar and communication systems for its ability of pulse compression and spread spectrum. This paper presents a principle of orthogonally multiplexing a bank of linear chirp waveforms within the same bandwidth. The amplitude and phase of the chirps are modulated for information communication. As Fourier trans-form is the basis for orthogonal frequency division multiplexing (OFDM), Fresnel transform underlies the proposed orthogonal chirp division multiplexing (OCDM). Digital implementa-tion of the OCDM system using discrete Fresnel transform is proposed. Based on the con-volution theorem of the Fresnel transform, the transmission of the OCDM signal is analyzed under the linear time-invariant or quasi-static channel with additive noise, which can gener-alize typical linear transmission channels. Based on the eigen-decomposition of Fresnel transform, efficient digital signal processing algorithm is proposed for compensating chan-nel dispersion by linear single- tap equalizers. The implementation details of the OCDM system is discussed with emphasis on its compatibility to the OFDM system. Finally, simula-tion are provided to validate the feasibility of the proposed OCDM under wireless channels. It is shown that the OCDM system is able to utilize the multipath diversity and outperforms the OFDM system under the multipath fading channels.

cs.IT↗

arXiv2015

Heterogeneous Cellular Networks Using Wireless Backhaul: Fast Admission Control and Large System Analysis

We consider a heterogeneous cellular network with densely underlaid small cell access points (SAPs). Wireless backhaul provides the data connection from the core network to SAPs. To serve as many SAPs and their corresponding users as possible with guaranteed data rates, admission control of SAPs needs to be performed in wireless backhaul. Such a problem involves joint design of transmit beamformers, power control, and selection of SAPs. In order to tackle such a difficult problem, we apply $\ell_1$-relaxation and propose an iterative algorithm for the $\ell_1$-relaxed problem. The selection of SAPs is made based on the outputs of the iterative algorithm. This algorithm is fast and enjoys low complexity for small-to-medium sized systems. However, its solution depends on the actual channel state information, and resuming the algorithm for each new channel realization may be unrealistic for large systems. Therefore, we make use of random matrix theory and also propose an iterative algorithm for large systems. Such a large system iterative algorithm produces asymptotically optimum solution for the $\ell_1$-relaxed problem, which only requires large-scale channel coefficients irrespective of the actual channel realization. Near optimum results are achieved by our proposed algorithms in simulations.

cs.IT↗

arXiv2014

Realization and design of a pilot assist decision-making system based on speech recognition

A system based on speech recognition is proposed for pilot assist decision-making. It is based on a HIL aircraft simulation platform and uses the microcontroller SPCE061A as the central processor to achieve better reliability and higher cost-effect performance. Technologies of LPCC (linear predictive cepstral coding) and DTW (Dynamic Time Warping) are applied for isolated-word speech recognition to gain a smaller amount of calculation and a better real-time performance. Besides, we adopt the PWM (Pulse Width Modulation) regulation technology to effectively regulate each control surface by speech, and thus to assist the pilot to make decisions. By trial and error, it is proved that we have a satisfactory accuracy rate of speech recognition and control effect. More importantly, our paper provides a creative idea for intelligent human-computer interaction and applications of speech recognition in the field of aviation control. Our system is also very easy to be extended and applied.

cs.HC↗

arXiv2014

1-bit Compressive Data Gathering for Wireless Sensor Networks

Compressive sensing (CS) has been widely used for the data gathering in wireless sensor networks for the purpose of reducing the communication overhead recent years. In this paper, we first show that with simple modification, 1-bit compressive sensing can also been used for the data gathering in wireless sensor networks to further reduce the communication overhead. We also propose a novel blind 1-bit CS reconstruction algorithm which outperforms other state of the art blind 1-bit CS reconstruction algorithms. Experimental results on real sensor datasets demonstrate the efficiency of our method.

cs.NI↗

arXiv2013

Cost Minimization in Multiple IaaS Clouds: A Double Auction Approach

IaaS clouds invest substantial capital in operating their data centers. Reducing the cost of resource provisioning, is their forever pursuing goal. Computing resource trading among multiple IaaS clouds provide a potential for IaaS clouds to utilize cheaper resources to fulfill their jobs, by exploiting the diversities of different clouds' workloads and operational costs. In this paper, we focus on studying the IaaS clouds' cost reduction through computing resource trading among multiple IaaS clouds. We formulate the global cost minimization problem among multiple IaaS clouds under cooperative scenario where each individual cloud's workload and cost information is known. Taking into consideration jobs with disparate lengths, a non-preemptive approximation algorithm for leftover job migration and new job scheduling is designed. Given to the selfishness of individual clouds, we further design a randomized double auction mechanism to elicit clouds' truthful bidding for buying or selling virtual machines. We evaluate our algorithms using trace-driven simulations.

cs.NI↗

arXiv2013

A quantum oscillator model for entropic gravity

In Verlinde's theory of entropic gravity, the internal degrees of freedom are marked by "bits", but the microscopic origin of "bits" is unknown hitherto. To this direction, we adopt the quantum harmonic oscillators to naively model the internal degrees of freedom. In this model, the equipartition relation is modified at the low temperature. The Newton's law and Poisson's equation of gravity are reproduced at small distances, but they receive appreciable corrections at large distances. Geometrization of gravity for this model is explored, resulting in modifications to Einstein equations.

hep-th↗

arXiv2013

Efficient Data Gathering in Wireless Sensor Networks Based on Matrix Completion and Compressive Sensing

Gathering data in an energy efficient manner in wireless sensor networks is an important design challenge. In wireless sensor networks, the readings of sensors always exhibit intra-temporal and inter-spatial correlations. Therefore, in this letter, we use low rank matrix completion theory to explore the inter-spatial correlation and use compressive sensing theory to take advantage of intra-temporal correlation. Our method, dubbed MCCS, can significantly reduce the amount of data that each sensor must send through network and to the sink, thus prolong the lifetime of the whole networks. Experiments using real datasets demonstrate the feasibility and efficacy of our MCCS method.

cs.NI↗

arXiv2012

Vertices, Vortices & Interacting Surface Operators

We show that the vortex moduli space in non-abelian supersymmetric N=(2,2) gauge theories on the two dimensional plane with adjoint and anti-fundamental matter can be described as an holomorphic submanifold of the instanton moduli space in four dimensions. The vortex partition functions for these theories are computed via equivariant localization. We show that these coincide with the field theory limit of the topological vertex on the strip with boundary conditions corresponding to column diagrams. Moreover, we resum the field theory limit of the vertex partition functions in terms of generalized hypergeometric functions formulating their AGT dual description as interacting surface operators of simple type. Analogously we resum the topological open string amplitudes in terms of q-deformed generalized hypergeometric functions proving that they satisfy appropriate finite difference equations.

hep-th↗

arXiv2012

Orbifold Vortex and Super Liouville Theory

We study the nonabelian vortex counting problem on $\mathbb{C}/\mathbb{Z}_p$. At first we calculate vortex partition functions on the orbifold space using localization techniques, then we find how to extract orbifold vortex partitions function from orbifold linear quiver instanton partition functions. Finally, we study the AGT like relation between orbifold SU(2) vortices and $\mathcal{N} = 1$ super Liouville theory in the mixed R/NS sector by fixing the dictionary among parameters in the common hypergeometric functions system.

hep-th↗

arXiv2011

The Liouville side of the Vortex

We analyze conformal blocks with multiple (semi-)degenerate field insertions in Liouville/Toda conformal field theories an show that their vector space is fully reproduced by the four-dimensional limit of open topological string amplitudes on the strip with generic boundary conditions associated to a suitable quiver gauge theory. As a byproduct we identify the non-abelian vortex partition function with a specific fusion channel of degenerate conformal blocks.

hep-th↗

arXiv2011

Adsorption and dissociation of water on Zr(0001) with density-functional theory studies

The adsorption and dissociation of isolated water molecule on Zr(0001) surface are theoretically investigated for the first time by using density-functional theory calculations. Two kinds of adsorption configurations with almost the same adsorption energy are identified as the locally stable states, i.e., the flat and upright configurations respectively. It is shown that the flat adsorption states on the top site are dominated by the 1$b_{1}$-$d$ band coupling, insensitive to the azimuthal orientation. The diffusion between adjacent top sites reveals that the water molecule is very mobile on the surface. For the upright configuration, we find that besides the contribution of the molecular orbitals 1$b_{1}$ and 3$a_{1}$, the surface$\rightarrow$water charge transfer occurring across the Fermi level also plays an important role. The dissociation of H$_{2}$O is found to be very facile, especially for the upright configuration, in good accordance with the attainable experimental results. The present results afford to provide a guiding line for deeply understanding the water-induced surface corrosion of zirconium.

cond-mat.mtrl-sci↗

arXiv2011

Adsorption and diffusion of H2O molecule on the Be(0001) surface: A density-functional theory study

Using first-principles calculations, we systematically study the adsorption behavior of a single molecular H$_{2}$O on the Be(0001) surface. We find that the favored molecular adsorption site is the top site with an adsorption energy of about 0.3 eV, together with the detailed electronic structure analysis, suggesting a weak binding strength of the H$_{2}$O/Be(0001) surface. The adsorption interaction is mainly contributed by the overlapping between the $s$ and $p_{z}$ states of the top-layer Be atom and the molecular orbitals 1$b_{1}$ and 3$a_{1}$ of H$_{2}$O. The activation energy for H$_{2}$O diffusion on the surface is about 0.3 eV. Meanwhile, our study indicates that no dissociation state exists for the H$_{2}$O/Be(0001) surface.

cond-mat.mtrl-sci↗

arXiv2010

Matrix Model and Refined Wall-Crossing Formula

In this paper, we show how to get matrix models corresponding to the refined BPS states partition functions of $\mathbb{C}^3$, resolved conifold and $\mathbb{C}^3/\mathbb{Z}_2$ by inserting the identity operator at a proper position in the fermionic expression of the refined BPS states partition functions.

hep-th↗