arXiv subjects

Yong Zhou

Publications and source records attributed to Yong Zhou.

At least 145 records · Page 8Linked to original sources

arXiv2020

Communication-Efficient Edge AI Inference Over Wireless Networks

Given the fast growth of intelligent devices, it is expected that a large number of high-stake artificial intelligence (AI) applications, e.g., drones, autonomous cars, tactile robots, will be deployed at the edge of wireless networks in the near future. As such, the intelligent communication networks will be designed to leverage advanced wireless techniques and edge computing technologies to support AI-enabled applications at various end devices with limited communication, computation, hardware and energy resources. In this article, we shall present the principles of efficient deployment of model inference at network edge to provide low-latency and energy-efficient AI services. This includes the wireless distributed computing framework for low-latency device distributed model inference as well as the wireless cooperative transmission strategy for energy-efficient edge cooperative model inference. The communication efficiency of edge inference systems is further improved by building up a smart radio propagation environment via intelligent reflecting surface.

eess.SP↗

arXiv2020

Federated Machine Learning for Intelligent IoT via Reconfigurable Intelligent Surface

Intelligent Internet-of-Things (IoT) will be transformative with the advancement of artificial intelligence and high-dimensional data analysis, shifting from "connected things" to "connected intelligence". This shall unleash the full potential of intelligent IoT in a plethora of exciting applications, such as self-driving cars, unmanned aerial vehicles, healthcare, robotics, and supply chain finance. These applications drive the need of developing revolutionary computation, communication and artificial intelligence technologies that can make low-latency decisions with massive real-time data. To this end, federated machine learning, as a disruptive technology, is emerged to distill intelligence from the data at network edge, while guaranteeing device privacy and data security. However, the limited communication bandwidth is a key bottleneck of model aggregation for federated machine learning over radio channels. In this article, we shall develop an over-the-air computation based communication-efficient federated machine learning framework for intelligent IoT networks via exploiting the waveform superposition property of a multi-access channel. Reconfigurable intelligent surface is further leveraged to reduce the model aggregation error via enhancing the signal strength by reconfiguring the wireless propagation environments.

eess.SP↗

arXiv2020

Reconfigurable Intelligent Surface Empowered Downlink Non-Orthogonal Multiple Access

Power-domain non-orthogonal multiple access (NOMA) has become a promising technology to exploit the new dimension of the power domain to enhance the spectral efficiency of wireless networks. However, most existing NOMA schemes rely on the strong assumption that users' channel gains are quite different, which may be invalid in practice. To unleash the potential of power-domain NOMA, we will propose a reconfigurable intelligent surface (RIS)-empowered NOMA network to introduce desirable channel gain differences among the users by adjusting the phase shifts at RIS. Our goal is to minimize the total transmit power by jointly optimizing the beamforming vectors at the base station and the phase-shift matrix at the RIS. To address the highly coupled optimization variables, we present an alternating optimization framework to decompose the non-convex bi-quadratically constrained quadratic problem into two rank-one constrained matrices optimization problems via matrix lifting. At the same time, to accurately detect the feasibility of the non-convex rank-one constraints and improve performance by avoiding early stopping in the alternating optimization procedure, we equivalently represent the rank-one constraint as the difference between nuclear norm and spectral norm. A difference-of-convex (DC) algorithm is further developed to solve the resulting DC programs via successive convex relaxation, followed by establishing the convergence of the proposed DC-based alternating optimization method. We further propose an efficient user ordering scheme with closed-form expressions, considering both the channel conditions and users' target data rates. Simulation results will validate the effectiveness of deploying an RIS and the superiority of the proposed DC-based alternating optimization method in reducing the total transmit power.

eess.SP↗

arXiv2020

Age of Information for Multicast Transmission with Fixed and Random Deadlines in IoT Systems

In this paper, we consider the multicast transmission of a real-time Internet of Things (IoT) system, where an access point (AP) transmits time-stamped status updates to multiple IoT devices. Different from the existing studies that only considered multicast transmission without deadlines, we enforce a deadline for the service time of each multicast status update, taking into account both the fixed and randomly distributed deadlines. In particular, a status update is dropped when either its deadline expires or it is successfully received by a certain number of IoT devices. Considering deadlines is important for many emerging IoT applications, where the outdated status updates are of no use to IoT devices. We evaluate the timeliness of the status update delivery by applying a recently proposed metric, named the age of information (AoI), which is defined as the time elapsed since the generation of the most recently received status update. After deriving the distributions of the service time for all possible reception outcomes at IoT devices, we manage to obtain the closed-form expressions of both the average AoI and the average peak AoI. Simulations validate the performance analysis, which reveals that the multicast transmission with deadlines achieves a lower average AoI than that without deadlines and there exists an optimal value of the deadline that can minimize the average (peak) AoI. Results also show that the fixed and random deadlines have respective advantages in different deadline regimes.

cs.IT↗

arXiv2020

On the Age of Information for Multicast Transmission with Hard Deadlines in IoT Systems

We consider the multicast transmission of a real-time Internet of Things (IoT) system, where a server transmits time-stamped status updates to multiple IoT devices. We apply a recently proposed metric, named age of information (AoI), to capture the timeliness of the information delivery. The AoI is defined as the time elapsed since the generation of the most recently received status update. Different from the existing studies that considered either multicast transmission without hard deadlines or unicast transmission with hard deadlines, we enforce a hard deadline for the service time of multicast transmission. This is important for many emerging multicast IoT applications, where the outdated status updates are useless for IoT devices. Specifically, the transmission of a status update is terminated when either the hard deadline expires or a sufficient number of IoT devices successfully receive the status update. We first calculate the distributions of the service time for all possible reception outcomes at IoT devices, and then derive a closed-form expression of the average AoI. Simulations validate the performance analysis, which reveals that: 1) the multicast transmission with hard deadlines achieves a lower average AoI than that without hard deadlines; and 2) there exists an optimal value of the hard deadline that minimizes the average AoI.

cs.IT↗

arXiv2020

Age-Oriented Opportunistic Relaying in Cooperative Status Update Systems with Stochastic Arrivals

This paper considers a cooperative status update system with a source aiming to send randomly generated status updates to a designated destination as timely as possible with the help of a relay. We adopt a recently proposed concept, Age of Information (AoI), to characterize the timeliness of the status updates. We propose an age-oriented opportunistic relaying (AoR) protocol to reduce the AoI of the considered system. Specifically, the relay opportunistically replaces the source to retransmit the successfully received status updates that have not been correctly delivered to the destination, but the retransmission of the relay can be preempted by the arrival of a new status update at the source. By carefully analyzing the evolution of AoI, we derive a closed-form expression of the average AoI for the proposed AoR protocol. We further minimize the average AoI by optimizing the generation probability of the status updates at the source. Simulation results validate our theoretical analysis and demonstrate that the average AoI performance of the proposed AoR protocol is superior to that of the non-cooperative system.

cs.IT↗

arXiv2019

Reconfigurable Intelligent Surface for Green Edge Inference

Reconfigurable intelligent surface (RIS) as an emerging cost-effective technology can enhance the spectrum- and energy-efficiency of wireless networks. In this paper, we consider an RIS-aided green edge inference system, where the inference tasks generated from resource-limited mobile devices (MDs) are uploaded to and cooperatively performed at multiple resource-enhanced base stations (BSs). Taking into account both the computation and uplink/downlink transmit power consumption, we formulate an overall network power consumption minimization problem, which calls for the joint design of the set of tasks performed by each BS, transmit and receive beamforming vectors of the BSs, transmit power of the MDs, and uplink/downlink phase-shift matrices at the RIS. Such a problem is a mixed combinatorial optimization problem with nonconvex constraints and is highly intractable. To address the challenge of the combinatorial objective, a group sparse reformulation is proposed by exploiting the group sparsity structure of the beamforming vectors, while a block-structured optimization (BSO) approach is proposed to decouple the optimization variables. Finally, we propose a BSO with mixed $\ell_{1,2}$-norm and difference-of-convex-functions (DC) based three-stage framework to solve the problem, where the mixed $\ell_{1,2}$-norm is adopted to induce the group sparsity of beamforming vectors and DC is adopted to effectively handle the nonconvex rank-one constraint after matrix lifting. Numerical results demonstrate the supreme gain of deploying an RIS and confirm the effectiveness of the proposed algorithm over the baseline algorithms.

eess.SP↗

arXiv2019

Constraining the anisotropy of the Universe via Pantheon supernovae sample

We test the possible dipole anisotropy of a Finslerian cosmological model and other three dipole-modulated cosmological models, i.e., the dipole-modulated $\rmΛ$CDM, $w$CDM and Chevallier--Polarski--Linder (CPL) model by using the recently released Pantheon sample of SNe Ia. The Markov chain Monte Carlo (MCMC) method is used to explore the whole parameter space. We find that the dipole anisotropy is very weak in all cosmological models used. Although the dipole amplitudes of four cosmological models are consistent with zero within $1σ$ uncertainty, the dipole directions are close to the axial direction to the plane of SDSS subsample among Pantheon. It may imply that the weak dipole anisotropy in the Pantheon sample originates from the inhomogeneous distribution of the SDSS subsample. More homogeneous distribution of SNe Ia is necessary to constrain the cosmic anisotropy.

astro-ph.CO↗

arXiv2019

Existence and regularity results for terminal value problem for nonlinear fractional wave equations

We consider the terminal value problem (or called final value problem, initial inverse problem, backward in time problem) of determining the initial value, in a general class of time-fractional wave equations with Caputo derivative, from a given final value. We are concerned with the existence, regularity of solutions upon the terminal value. Under several assumptions on the nonlinearity, we address and show the well-posedness (namely, the existence, uniqueness, and continuous dependence) for the terminal value problem. Some regularity results for the mild solution and its derivatives of first and fractional orders are also derived. The effectiveness of our methods are showed by applying the results to two interesting models: Time fractional Ginzburg-Landau equation, and Time fractional Burgers equation, where time and spatial regularity estimates are obtained.

math.AP↗

arXiv2019

On existence and regularity of a terminal value problem for the time fractional diffusion equation

In this paper we consider a final value problem for a diffusion equation with time-space fractional differentiation on a bounded domain $D$ of $ \mathbb{R}^{k}$, $k\ge 1$, which includes the fractional power $\mathcal L^β$, $0<β\le 1$, of a symmetric uniformly elliptic operator $\mathcal L$ defined on $L^2(D)$. A representation of solutions is given by using the Laplace transform and the spectrum of $\mathcal L^β$. We establish some existence and regularity results for our problem in both the linear and nonlinear case.

math.AP↗

arXiv2019

Forecasting security's volatility using low-frequency historical data, high-frequency historical data and option-implied volatility

Low-frequency historical data, high-frequency historical data and option data are three major sources, which can be used to forecast the underlying security's volatility. In this paper, we propose two econometric models, which integrate three information sources. In GARCH-Itô-OI model, we assume that the option-implied volatility can influence the security's future volatility, and the option-implied volatility is treated as an observable exogenous variable. In GARCH-Itô-IV model, we assume that the option-implied volatility can not influence the security's volatility directly, and the relationship between the option-implied volatility and the security's volatility is constructed to extract useful information of the underlying security. After providing the quasi-maximum likelihood estimators for the parameters and establishing their asymptotic properties, we also conduct a series of simulation analysis and empirical analysis to compare the proposed models with other popular models in the literature. We find that when the sampling interval of the high-frequency data is 5 minutes, the GARCH-Itô-OI model and GARCH-Itô-IV model has better forecasting performance than other models.

q-fin.ST↗

arXiv2019

Intelligent Reflecting Surface for Downlink Non-Orthogonal Multiple Access Networks

Intelligent reflecting surface (IRS) has recently been recognized as a promising technology to enhance the energy and spectrum efficiency of wireless networks by controlling the wireless medium with the configurable electromagnetic materials. In this paper, we consider the downlink transmit power minimization problem for a IRS-empowered non-orthogonal multiple access (NOMA) network by jointly optimizing the transmit beamformers at the BS and the phase shift matrix at the IRS. However, this problem turns out to be a highly intractable non-convex bi-quadratic programming problem, for which an alternative minimization framework is proposed via solving the non-convex quadratic programs alternatively. We further develop a novel difference-of-convex (DC) programming algorithm to solve the resulting non-convex quadratic programs efficiently by lifting the quadratic programs into rank-one constrained matrix optimization problems, followed by representing the non-convex rank function as a DC function. Simulation results demonstrate the performance gains of the proposed method.

eess.SP↗

arXiv2019

Anisotropy of the Universe via the Pantheon supernovae sample revisited

We employ the hemisphere comparison (HC) method and the dipole fitting (DF) method to investigate the cosmic anisotropy in the recently released Pantheon sample of type Ia supernovae (SNe Ia) and five combinations among Pantheon. For the HC method, we find the maximum anisotropy level in the full Pantheon sample is $\mathrm{AL}_{max}=0.361\pm0.070$ and corresponding direction $(l,b)=({123.05^{\circ}}^{+11.25^{\circ}}_{-4.22^{\circ}}, {4.78^{\circ}}^{+1.80^{\circ}}_{-8.36^{\circ}})$. A robust check shows the statistical significance of maximum anisotropy level is about $2.1σ$. We also find that the Low-$z$ and SNLS subsamples have decisive impact on the overall anisotropy while other three subsamples have little impact. Moreover, the anisotropy level map significantly rely on the inhomogeneous distribution of SNe Ia in the sky. For the DF method, we find the dipole anisotropy in the Pantheon sample is very weak. The dipole magnitude is constrained to be less than $1.16\times10^{-3}$ at $95\%$ confidence level. However, the dipole direction is well inferred by MCMC method and it points towards $(l,b)=({306.00^{\circ}}^{+82.95^{\circ}}_{-125.01^{\circ}}, {-34.20^{\circ}}^{+16.82^{\circ}}_{-54.93^{\circ}})$. This direction is very close to the axial direction to the plane of SDSS subsample. It may imply that SDSS subsample is the decisive part to the dipole anisotropy in the full Pantheon sample. All these facts imply that the cosmic anisotropy found in Pantheon sample significantly rely on the inhomogeneous distribution of SNe Ia in the sky. More homogeneous distribution of SNe Ia is necessary to search for a more convincing cosmic anisotropy.

astro-ph.CO↗

arXiv2019

Video Object Segmentation and Tracking: A Survey

Object segmentation and object tracking are fundamental research area in the computer vision community. These two topics are diffcult to handle some common challenges, such as occlusion, deformation, motion blur, and scale variation. The former contains heterogeneous object, interacting object, edge ambiguity, and shape complexity. And the latter suffers from difficulties in handling fast motion, out-of-view, and real-time processing. Combining the two problems of video object segmentation and tracking (VOST) can overcome their respective difficulties and improve their performance. VOST can be widely applied to many practical applications such as video summarization, high definition video compression, human computer interaction, and autonomous vehicles. This article aims to provide a comprehensive review of the state-of-the-art tracking methods, and classify these methods into different categories, and identify new trends. First, we provide a hierarchical categorization existing approaches, including unsupervised VOS, semi-supervised VOS, interactive VOS, weakly supervised VOS, and segmentation-based tracking methods. Second, we provide a detailed discussion and overview of the technical characteristics of the different methods. Third, we summarize the characteristics of the related video dataset, and provide a variety of evaluation metrics. Finally, we point out a set of interesting future works and draw our own conclusions.

cs.CV↗

arXiv2019

iQIYI-VID: A Large Dataset for Multi-modal Person Identification

Person identification in the wild is very challenging due to great variation in poses, face quality, clothes, makeup and so on. Traditional research, such as face recognition, person re-identification, and speaker recognition, often focuses on a single modal of information, which is inadequate to handle all the situations in practice. Multi-modal person identification is a more promising way that we can jointly utilize face, head, body, audio features, and so on. In this paper, we introduce iQIYI-VID, the largest video dataset for multi-modal person identification. It is composed of 600K video clips of 5,000 celebrities. These video clips are extracted from 400K hours of online videos of various types, ranging from movies, variety shows, TV series, to news broadcasting. All video clips pass through a careful human annotation process, and the error rate of labels is lower than 0.2\%. We evaluated the state-of-art models of face recognition, person re-identification, and speaker recognition on the iQIYI-VID dataset. Experimental results show that these models are still far from being perfect for the task of person identification in the wild. We proposed a Multi-modal Attention module to fuse multi-modal features that can improve person identification considerably. We have released the dataset online to promote multi-modal person identification research.

cs.CV↗

arXiv2019

Is there a fundamental acceleration scale in galaxies?

Milgrom's modified Newtonian dynamics (MOND) can explain well the mass discrepancy problem in galaxy without invoking dark matter. The MOND theory predicts a universal constant acceleration scale in galaxy, below which the Newtonian dynamics is no longer valid. Recently, Rodrigues et al. analyzed 193 high-quality disk galaxies by using Bayesian inference and they concluded that the probability of existence of a fundamental acceleration scale is essentially 0. In this paper, we use the same method to revisit the same question: Is there a fundamental acceleration scale in galaxies? We fit each galaxy rotation curve of 175 SPARC galaxies with Gaussian priors on galaxy parameters and a variable acceleration scale. After marginalizing over the nuisance parameters, we find that the marginalized posterior distributions of the acceleration scale become broad. The incompatibility between the global best fit and the marginalized posterior distributions of the acceleration scale is greatly reduced. However, there still exists evidence that rejects MOND as a fundamental theory. More accurate observations are needed to exclude or confirm the existence of a fundamental acceleration scale in galaxies.

astro-ph.GA↗

arXiv2019

Proportional Mean Residual Life Model with Censored Survival Data under Case-cohort Design

Proportional mean residual life model is studied for analysing survival data from the case-cohort design. To simultaneously estimate the regression parameters and the baseline mean residual life function, weighted estimating equations based on an inverse selection probability are proposed. The resulting regression coefficients estimates are shown to be consistent and asymptotic normal with easily estimated variance-covariance. Simulation studies show that the proposed estimators perform very well. An application to a real dataset from the South Welsh nickel refiners study is also given to illustrate the methodology.

math.ST↗

arXiv2018

Miniaturized high-frequency sine wave gating InGaAs/InP single-photon detector

High-frequency gating InGaAs/InP single-photon detectors (SPDs) are widely used for applications requiring single-photon detection in the near-infrared region such as quantum key distribution. Reducing SPD size is highly desired for practical use, which is favorable to the implementation of further system integration. Here we present, to the best of our knowledge, the most compact high-frequency sine wave gating (SWG) InGaAs/InP SPD. We design and fabricate an InGaAs/InP single-photon avalanche diode (SPAD) with optimized semiconductor structure, and then encapsulate the SPAD chip and a mini-thermoelectric cooler inside a butterfly package with a size of 12.5 mm $\times$ 22 mm $\times$ 10 mm. Moreover, we implement a monolithic readout circuit for the SWG SPD in order to replace the quenching electronics that is previously designed with board-level integration. Finally, the components of SPAD, monolithic readout circuit and the affiliated circuits are integrated into a single module with a size of 13 cm $\times$ 8 cm $\times$ 4 cm. Compared with the 1.25 GHz SWG InGaAs/InP SPD module (25 cm $\times$ 10 cm $\times$ 33 cm) designed in 2012, the volume of our miniaturized SPD is reduced by 95\%. After the characterization, the SPD exhibits excellent performance with a photon detection efficiency of 30\%, a dark count rate of 2.0 kcps and an afterpulse probability of 8.8\% under the conditions of 1.25 GHz gating rate, 100 ns hold-off time and 243 K. Also, we perform the stability test over one week, and the results show the high reliability of the miniaturized SPD module.

physics.ins-det↗