A5101493939- Geophysics and Gravity Measurements
- GNSS positioning and interference
- UAV Applications and Optimization
- Infrared Target Detection Methodologies
- Smart Grid Energy Management
- Statistical Methods and Inference
- Robotic Path Planning Algorithms
- Vehicular Ad Hoc Networks (VANETs)
- Radio Wave Propagation Studies
- Machine Learning and Algorithms
- Energy Load and Power Forecasting
- Sparse and Compressive Sensing Techniques
- Ionosphere and magnetosphere dynamics
- Tensor decomposition and applications
- Millimeter-Wave Propagation and Modeling
- Advanced biosensing and bioanalysis techniques
- Spectroscopy and Chemometric Analyses
- Distributed Sensor Networks and Detection Algorithms
- Gene expression and cancer classification
- Multi-Criteria Decision Making
- Electricity Theft Detection Techniques
- Direction-of-Arrival Estimation Techniques
- Distributed Control Multi-Agent Systems
- Advanced Electron Microscopy Techniques and Applications
- Advanced Optimization Algorithms Research
National Taiwan University
2023
National Central University
2020-2022
University of Illinois Urbana-Champaign
2008-2016
General Electric (United States)
2015-2016
GE Global Research (United States)
2015
Southwest Jiaotong University
2007-2008
This article concerns the problem of trajectory design and base station (BS) association for cellular-connected unmanned aerial vehicles (UAVs). To support safety-critical functions, one primary requirement UAVs is to maintain reliable cellular connectivity at every time instant during flight mission. Since antenna gain a ground BS (GBS) changes with position UAV, UAV-GBS strategy should be jointly considered design, which has not been studied in prior arts. In this article, we first...
This paper studies the trajectory optimization problem of a single cellular-enabled unmanned aerial vehicle (UAV), taking into account outage performance entire trajectory. To provide real-time control, it is critical for UAV to maintain reliable connectivity with ground base station (GBS). We first consider performance, which defined as sum time duration not meeting predefined threshold during mission. Then we formulate minimize mission completion time, while ensuring constraint...
The deployment of smart meters has made available high-frequency (minutes as opposed to monthly) measurements electricity usage at individual households. Converting these knowledge that can improve energy efficiency in the residential sector is critical attract further grid investments and engage consumers path towards reducing global carbon footprint. goal reported research use meter measurement data identify heating cooling levels for a home. This important cost effectively design consumer...
In recent years, solutions to various hypothesis testing problems in the asymptotic setting have been proposed using results from large deviation theory. Such tests are optimal terms of appropriately defined error exponents. For practitioner, however, probabilities finite sample size more important. this paper, we show how on weak convergence test statistic can be used obtain better approximations for setting. While technique is popular among statisticians common tests, demonstrate its...
This paper is concerned with the model reduction for Markov chain models. The goal to obtain a low-rank approximation original transition matrix. A nuclear-norm regularized optimization problem proposed this purpose, in which Kullback-Leibler divergence rate used measure similarity between two chains, and nuclear norm approximate rank function. An efficient iterative algorithm developed compute solution problem. effectiveness of approach demonstrated via numerical examples.
We introduce a new algebraic framework for detecting spot failures in DNA microarrays. The technique leverages the theory of superimposed coding with iterative detection methods, and has advantage being constructive small implementation complexity, as opposed to existing approaches microarray error-control coding.
Many recent works have proposed spatial channel models for MIMO applications in the 1 to 10 GHz band. However, attention has shifted 60 regime where opening up of spectrum created new opportunities multi-antenna communications. With a broad focus on regime, we revisit modeling classical setting and propose an angular domain decomposition this work. In special case non-uniform linear arrays, proposal corresponds non-uniformly partitioning commensurate with non-uniformity structure arrays...
This paper considers the problem of feature selection for composite hypothesis testing: The goal is to select, from m candidate features, r relevant ones distinguishing null alternative hypothesis; training data are given as L sequences observations, which each an n-sample sequence coming one distribution in hypothesis. What fundamental limit successful selection? Are there any algorithms that achieve this limit? We investigate a small-sample high-dimensional setting, with n = o(m), and...
The ongoing smart grid transformation in utility networks is making available fine grained measurements of electricity consumption. To realize the full potential collected data we apply sophisticated analytics and machine learning techniques to correlate consumption with other types demographic (household surveys tax records) place within right context. This context setting achieved by a rigorous feature selection procedure, followed clustering group customers into peer groups. statistical...
Single source images do not contain the comprehensive information of target and recognition rate is low. Targeting to these problems, a new method based on fusing features in Visible (VIS), Middle Wave Infrared (MWIR) Long (LWIR) proposed. First, typical VIS, MWIR LWIR are extracted. The fused feature vector subsequently obtained by selection Fisher discriminant analysis. Finally, realized through Support Vector Machine (SVM). performance proposed five other algorithms were tested compared...
In this paper, we study the performance of cellular-connected unmanned aerial vehicles (UAVs) in context an intelligent ground base station (GBS) association scheme. We aim to maintain reliable wireless communication links between moving UAVs and GBSs. Given beneficial line-of-sight (LoS) propagation, a UAV may have more candidate GBSs for compared that terrestrial networks. By exploiting radio map objective area, deep neural network (DNN) based scheme is proposed determining optimal GBS...
<abstract> <b><sc>Abstract.</sc></b> In order to seek a rapid detection of delinted cottonseed vigor without destroying, hyperspectral imaging technology was investigated detect the in this study. Experiment with different aging degree three varieties cottonseeds as research object, such Xin Luzao 50, 52, 62. Firstly, image range 450-1013nm collected. Secondly, we extracted 720 RIOs and each region interest contain single entire cottonseeds. Thirdly, extraction spectral data used...
This paper is concerned with model reduction for Markov chain models. The goal to obtain a low-rank approximation the original chain. Kullback–Leibler divergence rate used measure similarity between two chains; nuclear norm approximate rank function. A nuclear-norm regularised optimisation problem formulated approximately find optimal approximation. proposed analysed and performance bounds are obtained through convex analysis. An iterative fixed point algorithm developed based on proximal...