Search

In this paper, we investigate the electrical characteristics of AlGaN/GaN HEMTs at the lowest temperature of 20 K. The measurement results indicate that the output current of the device decreases significantly with increasing temperature at temperature ranging from 40 K to 260 K, and the saturation drain current decreases by 19%. The gate leakage current rises slightly when the temperature increases. However, both the transfer and C-V characteristics indicate that the threshold voltage shift slightly in a negative direction as the temperature rises. In order to determine the physical mechanism of electrical characteristics change, the low-frequency noise (LFN) characteristics at different temperatures were measured and the density of traps was extracted. Finally, we consider that there are two competing mechanisms affecting the electrical characteristics of devices. The trap density reduction caused by temperature rise leads to threshold voltage’s negative shift, while the drop of 2DEG mobility is the main reason for the decrease of output current.

It is important in production to achieve accurate counting and density estimation of high-density culture fry under the environmental conditions of aquaculture scenarios in an efficient and accurate manner. However, none of the current methods for fry counting works well under the high-density and high-overlap conditions of real aquaculture scenarios. Therefore, in this paper, we propose a high-density farming fry monitoring network model, Super-Resolution GAN Density Estimate Attention Network (SGDAN), which incorporating an image enhancement algorithm and an attention mechanism, and we create a high-density farming fry dataset (HD-FryDataset) based on the environmental conditions of real aquaculture scenarios. The network model is designed to improve and optimize the targeted subnetworks for several key aspects of high-density fish fry monitoring work. Four subnetworks are included for image optimization, feature extraction, attention, and density map estimation. The experimental results show that the SGDAN network model achieved an average counting accuracy of 97.57% on the high-density culture fry dataset, which was 8.23% and 2.06% higher than those of MCNN and CSRNet, respectively. Additionally, the MAE and RMSE of the model were reduced by 71.9% and 67.3% and by 34.3% and 33.2% compared with those of MCNN and CSRNet, respectively. The model proposed in this paper also has a better ability to generate predictive density maps. The density maps generated by SGDAN have values of the evaluation metrics PSNR and SSIM of 20.33 and 0.933, respectively, which are 3.31 and 0.037 and 2.63 and 0.031 higher than those of MCNN and CSRNet. In general, the network model proposed in this paper outperforms existing network models in two applications: accurate counting of fry and generation of density maps for high-density culture in aquaculture. It also provides a good solution for digitizing the number of fry and visualizing the density of high-density culture in intelligent aquaculture systems.

Small object detection is one of the main obstacles hindering the development of object detection technology. In small object detection tasks, the performance of universal object detectors often drops sharply. We find through experiments on common models that for small objects, the box prior based on anchor detectors and the point prior without anchor detectors are suboptimal. The current anchor based or anchor-free label allocation methods will generate many false positive small objects samples, resulting in a decrease in detector attention to small objects. Therefore, we propose a new detector CLAHR for small object detection. In response to the issue of high sensitivity to small objects using Intersection over Union (IoU), where small deviations can lead to poor quality of assigned positive and negative samples, CLAHR utilizes the insensitivity of Gaussian distribution to small perturbations of small objects to transform the predicted box and manually annotated true box into Gaussian distribution. Then, CLAHR uses Wasserstein Distance (WD) to measure the distance between the predicted actual receptive field (ARF) and the manually labeled true receptive field (TRF) of the real box, rather than using IoU or central sampling strategies to allocate samples. Considering that IoU threshold based and central sampling strategies tend to favor large objects, we further design a cascaded label assignment (CLA) module to achieve balanced learning for small objects samples. Extensive experiments on the COCO, AI-TOD, and VisDrone datasets demonstrate the effective-eness of the proposed approach.

This study mainly addresses the problem of spatial information network link congestion and how to ensure the quality of user experience, while improving the accuracy and interpretability of decision-making, making it more suitable for dealing with complex network environments. It proposes an intelligent spatial routing strategy on the ground of convolutional neural networks and fuzzy logic. Firstly, a time-varying graph model is established to describe the time-varying characteristics of satellite nodes and inter satellite links, and the processed physical and state data is used as input to the model. The results showed that the research strategy performed well in terms of processing power, recognition accuracy, and time complexity. The cost gap between the methods proposed by the research institute and intrusion detection systems is gradually becoming apparent and expanding. Specifically, the growth rate of the methods proposed in the study is significantly slower than that of intrusion detection systems, at 50MB, 60MB, 78MB, 82MB, and 98MB, respectively. In network performance analysis, the research strategy shows excellent network traffic balancing ability by minimizing network congestion values. Meanwhile, the research strategy performs excellently in resource utilization efficiency, is not affected by changes in network topology, and has unique advantages. From this, it can be seen that research strategies have shown excellent performance in handling large-scale tasks, network congestion management, resource utilization efficiency, computing speed and scalability, and controlling signaling overhead; This demonstrates its important application value in the integrated communication system of heaven and earth.

The deployment of dynamic reactive power source can effectively improve the voltage performance after a disturbance for a power system with increasing wind power penetration level and ubiquitous induction loads. To improve the voltage stability of the power system, this paper proposes an adaptive many-objective robust optimization model to deal with the deployment issue of dynamic reactive power sources. Firstly, two metrics are adopted to assess the voltage stability of the system at two different stages, and one metric is proposed to assess the tie-line reactive power flow. Then, a robustness index is developed to assess the sensitivity of a solution when subjected to operational uncertainties, using the estimation of acceptable sensitivity region (ASR) and D-vine Copula. Five objectives are optimized simultaneously: ① total equipment investment; ② adaptive short-term voltage stability evaluation; ③ tie-line power flow evaluation; ④ prioritized steady-state voltage stability evaluation; and ⑤ robustness evaluation. Finally, an angle-based adaptive many-objective evolutionary algorithm (MaOEA) is developed with two improvements designed for the application in a practical engineering problem: ① adaptive mutation rate; and ② elimination procedure without a requirement for a threshold value. The proposed model is verified on a modified Nordic 74-bus system and a real-world power system. Numerical results demonstrate the effectiveness and efficiency of the proposed model.

The penetration of distributed energy resources (DERs) and energy-intensive resources is gradually increasing in active distribution networks (ADNs), which leads to frequent and severe voltage violation problems. As a densely distributed flexible resource in the future distribution network, 5G base station (BS) backup battery is used to regulate the voltage profile of ADN in this paper. First, the dispatchable potential of 5G BS backup batteries is analyzed. Considering the spatial-temporal characteristics of electric load for 5G BS, the dispatchable capacity of backup batteries at different time intervals is evaluated based on historical heat map data. Then, a voltage profile optimization model for ADN is established, consisting of 5G BS backup batteries and other voltage regulation resources. In this model, the charging/discharging behavior of backup batteries is based on its evaluation result of dispatchable capacity. Finally, the range of charging/discharging cost coefficients of 5G BS that benefits ADN and 5G operators are analyzed respectively. Further, an incentive policy for 5G operators is proposed. Under this policy, the charging/discharging cost coefficients of 5G BS can achieve a win-win situation for ADN and 5G operators. As an emerging flexible resource in ADN, the effectiveness and economy of 5G BS backup batteries participating in voltage profile optimization are verified in a test distribution network.

The load in the power grid is often affected by many factors, and the coupling relationship among them changes dynamically with time. This study proposed a short-term load forecasting technique based on time pattern attention and a long short-term memory network considering time-dependent intrinsic cross-correlation (TDICC) for smart grids with massive amounts of data. Based on source-load dynamic correlation analysis, The proposed TDICC to track and correct the multi-timescale dynamic correlation of two signal overruns or lags, which realizes the transformation of correlation description from two-dimensional space to three-dimensional space and expands its ability to describe multi-timescale dynamic correlation. Finally, the actual load data are used for example analysis, and the results show that the proposed method can tap the dynamic correlation between multiple influencing factors and load and has higher prediction accuracy compared with other models, which provides a more accurate database for power system dispatching.

To optimize the placement of soft open points (SOPs) in active distribution networks (ADNs), many aspects should be considered, including the adjustment of transmission power, integration of distributed generations (DGs), coordination with conventional control methods, and maintenance of economic costs. To address this multi-objective planning problem, this study proposes a multi-stage coordinated robust optimization model for the SOP allocation in ADNs with photovoltaic (PV). First, two robust technical indices based on a robustness index are proposed to evaluate the operation conditions and robust optimality of the solutions. Second, the proposed coordinated allocation model aims to optimize the total cost, robust voltage offset index, robust utilization index, and voltage collapse proximity index. Third, the optimization methods of the multi-and single-objective models are coordinated to solve the proposed multi-stage problem. Finally, the proposed model is implemented on an IEEE 33-node distribution system to verify its effectiveness. Numerical results show that the proposed index can better reveal voltage offset conditions as well as the SOP utilization, and the proposed model outperforms conventional ones in terms of robustness of placement plans and total cost.

Accurate shrimp seed counting is an important task in the aquaculture and stocking shrimp seeds processes. At present, most shrimp seed counting tasks are still performed manually, which is not only time-consuming and laborious but also has low accuracy, and this counting method cannot realize modern aquaculture. Therefore, to promote the modernization of aquaculture, a portable shrimp seed counting system is proposed in this paper. The system is based on Shrimpseed_Net, a modified CSRNet convolutional neural network, and can achieve higher accuracy in shrimp seed counting. The algorithm has been successfully implemented on a smartphone, enabling automatic counting of shrimp seeds within seconds by simply taking a picture of the shrimp seeds or selecting an image from the photo album to upload to the internal algorithm. Experimental results show that after only 50 iterations, the average absolute error of the algorithm is reduced to 17.28, with an accuracy rate of 95.53%. The system can accurately count shrimp seeds in real scenarios within a short period. The shrimp seed counting system studied in this paper can be widely used in the aquaculture industry to promote its development and enable intelligent aquaculture practices.

Bipolar direct current (DC) distribution networks can effectively improve the connection flexibility for renewable generations and loads. In practice, concerns regarding the potential voltage unbalance issue of the distribution networks and the frequency of switching still remain. This paper proposes a day-ahead polarity switching strategy to reduce voltage unbalance by optimally switching the polarity of renewable generations and loads while minimizing the switching times simultaneously in the range of a full day. First, a multi-objective optimization model is constructed to minimize the weighted sum of voltage unbalance factors and the sum of number of switching actions in the day based on the power flow model. Second, a two-step solution strategy is proposed to solve the optimization model. Finally, the proposed strategy is validated using 11-node and 34-node distribution networks as case studies, and a switching and stabilizing device is designed to enable unified switching of renewable generations and loads. Numerical results demonstrate that the proposed strategy can effectively reduce the switching times without affecting the improvement of voltage balance.

This paper studies the phase and gain control dependencies of a variable gain amplifier (VGA) and a vector interpolator phase shifter (VIPS) within a 20 GHz beamforming receiver. First, the mechanisms of gain control and phase variation in a classic current-steering VGA are analyzed and design techniques are proposed such that the gain-dependent phase variations (GDPV) introduced by the amplifiers are well balanced. Second, similar analysis is performed to evaluate GDPV within a vector interpolator, where we show how the same techniques only partially apply due to the cross-coupling structure of the interpolator’s VGAs. We evaluate our techniques within a 20 GHz beamforming receiver IC realized in GlobalFoundries 45 nm RFSOI. Very low GDPV is observed within the VGA, with less than 0.3-deg. root-mean squared phase variation for a 9 dB gain control, whereas the VIPS achieves worst-case GDPV of 14 deg. The full beamformer channel achieves 29 dB gain, 2.2-2.4 dB noise figure, −26 dBm input 1 dB compression and consumes 111 mW. Based on these results, improvements to the interpolator are proposed and validated through simulation. The result is a near-ideal Cartesian interpolator that has less than 2-deg. GDPV and very low crosstalk between the VGAs.

The ratio of torque to copper loss can be increased by injecting third-harmonic current into five-phase permanent-magnetic synchronous motor (PMSM). Accurate flux linkage information is indispensable in the process of calculating third-harmonic current reference corresponding to the maximum torque/copper loss ratio. In harsh working environments, the flux linkage will deviate from the initial measured value, resulting in the deviation between the third-harmonic current and the maximum torque/copper working point. Therefore, a third-harmonic current injection control strategy of five-phase PMSM based on third-harmonic current reference online identification is proposed in this paper. The key of this strategy is to transform the steady-state voltage equations into an equation containing only voltages and currents. The transformed equation has a characteristic that it is equal to 0 when the motor operates at the maximum torque/copper loss ratio working point. Meanwhile, an observer is designed to estimate the third-harmonic current reference by using the characteristic. Finally, the observer is applied in third-harmonic current injection control of five-phase PMSM. Compared with previous methods, the proposed strategy is simple, no need of flux linkage information and not affected by the change of flux linkage. The experimental results demonstrate the robustness of the proposed strategy.

A bandwidth-tunable multiband microwave photonic filter (MPF) using a strongly coupled microring resonator (MRR) with optical phase modulation is demonstrated theoretically and experimentally. Two notches of the MRR adjacent to the optical carrier break the intrinsic balance of the phase-modulated signal and then two sub-MPFs are formed thanks to the conversion from phase modulation to intensity modulation (PM-IM). The frequency responses of the two sub-MPFs are superimposed to form a passband with a fixed center frequency. Due to the periodicity of the MRR transmission spectrum, multiple passbands separated by a certain spacing appear in the frequency domain and combine to form a multiband MPF, whose bandwidth can be adjusted by tuning the wavelength of the carrier. The tunable bandwidth and shape factors ranging from 0.73 GHz to 2.73 GHz and 5.06 to 1.38 are experimentally performed. The number of passbands can be reconfigured. The proposed multiband MPF has the potential to be employed in modern multi-standard wireless communication systems.

In northern China, thermal power units (TPUs) are important in improving the penetration level of renewable energy. In such areas, the potentials of coordinated dispatch of renewable energy sources (RESs) and TPUs can be better realized, if RESs and TPUs connected to the power grid at the same point of common coupling (PCC) are dispatched as a coupled system. Firstly, the definition of the coupled system is introduced, followed by an analysis on its characteristics. Secondly, based on the operation characteristics of deep peak regulation (DPR) of TPUs in the coupled system, the constraint of the ladder-type ramping rate applicable for day-ahead dispatch is proposed, and the corresponding flexible spinning reserve constraint is further established. Then, considering these constraints and peak regulation ancillary services, a day-ahead optimal dispatch model of the coupled system is established. Finally, the operational characteristics and advantages of the coupled system are analyzed in several case studies based on a real-world power grid in Liaoning province, China. The numerical results show that the coupled system can further improve the economic benefits of RESs and TPUs under the existing policies.

CryoSat-2 repeatedly collects dense radar altimetry footprints covering high latitudes of the polar ice sheets for over ten years. The Baseline-D height product of CryoSat-2 was recently upgraded and released in 2019. Based on the internal or locally external evaluation of CryoSat-2 Baseline-D ice heights, we extended the validation both heterogeneously and spatially to accomplish a comprehensive assessment. First, reliable ice surface global navigation satellite system point solutions along the 36th Chinese National Antarctic Research Expedition traverse functioned as a height reference for comparison with the CryoSat-2 SAR synthetic aperture radar interferometric mode (SIN) data. Secondly, the Land Ice Along-Track Height product over Lambert–Amery system (LAS) from ice, cloud, and land elevation satellite 2 (ICESat-2), was also applied to validate CryoSat-2 SIN and low-resolution mode (LRM) data. As the results indicate, the SIN height accuracy is evaluated to be −1.68 m ± 2.35 m (slope < 0.65°) along the traverse and −0.96 ± 1.95 m (slope < 0.95° and roughness < 3 m) over the margin of LAS. The LRM height accuracy is evaluated to be −0.13 ± 0.46 m (slope < 0.1° & roughness < 1 m) over the interior of LAS. The standard deviation of the height differences degrades linearly against the limited growth of the terrain slope/roughness at the confidence level of 99%. It is worth noting that the spatially heterogeneous pattern of height differences is correlated with surface topography variations. This article should imply the potential to improve the estimation of mass balance and its uncertainty of polar ice sheets based on radar altimetry data.

At the site of continuous rolling of lithium battery electrode, most of its correction units use traditional PID control algorithms, coupled with the complex structure of the rolling equipment and the large number of control system actuators, result in the deviation of the pole piece position being controlled within 10mm only. In order to avoid the undesirable industrial problems caused by untimely correction and insufficient control accuracy, this paper proposes a control strategy for lithium battery roll press deflection device with the introduction of GA(Genetic Algorithm) optimized integral separation PID. For the problems of poor self-adaptive capability of deflection control and interference with equipment and environment during the rolling process. Analyze the causes of runout of the pole piece and the structure of the deflection correction system, establish a mathematical model and an integral separation fuzzy PID controller to realize the deflection control. Optimize the key parameters of the controller using GA to improve the control accuracy and performance of the deflection control. Simulations and experiments show that the deviation control strategy proposed in this paper can reduce the position deviation of the polar strip to within 4mm, which can effectively improve the deviation correction accuracy and the anti-interference capability of equipment operation, and also provide a good solution for the deviation control of other large and complex mechanical equipment.

As a core step in clustering analysis, distance measurement results can influence clustering accuracy. Existing measurement methods are mostly based on cluster feature information. However, these cluster features may be insufficient and result in losing data information for clusters containing a number of objects. To improve measurement accuracy, we make full use of the distribution characteristics of objects in clusters, i.e., we use descriptive statistics and the Wilcoxon-Mann-Whitney rank sum test in nonparametric statistics to measure distances during clustering. Furthermore, we propose a two-stage clustering algorithm to improve clustering analysis performance. In terms of avoiding preliminarily assuming the number of clusters, with the proposed distance measurement method, the clustering algorithm can discover clusters with arbitrary shapes and improve clustering accuracy. Experiments with multiple datasets compared with other clustering algorithms illustrate the accuracy and efficiency of the proposed clustering algorithm.

Soil moisture (SM) is a key parameter of the hydrological process, which affects exchanges of water and heat at the land/atmosphere interface. The “trapezoid” (or “triangle”) method has been widely applied to SM monitoring based on the pixel distribution within the thermal and optical remote sensing observations. However, the trapezoid method is a linear empirical model highly related to the retrieval accuracy of the surface temperature. In the article, the moderate-resolution imaging spectroradiometer (MODIS) data were applied to retrieve SM through an improved method over the Tibetan Plateau. The improved method is integrated with the “trapezoid” model and multiple learning techniques, Random Forest (RF) and Extreme Gradient Boosting (XGBoost). Meanwhile, RF and XGBoost were both trained with SM target data (the scale of SM and soil temperature) derived from the Tibetan Plateau observations, and the input variables were derived from MODIS observations. Compared with the SM measured, the results showed the root mean square error, the mean absolute error, and the correlation coefficient of the ensemble retrievals were 0.046-0.081 m3m-3, 0.030-0.065 m3m-3, and 0.60-0.87, respectively, which is better than that of the separate model. The ideas to implement the combination of traditional inversion algorithms and machine learning methods are helpful for researches in remote sensing fields.

Since the decision trees (DTs) have an advantage over “black-box” models, such as neural nets or support vector machines, in terms of comprehensibility, such that it might merit improvement for further optimization. The node splitting measures and pruning methods are primary among the techniques that can improve the generalization abilities of DTs. Here, we introduced the unequal interval optimization for node splitting, as well as the local chi-square test for tree pruning. This new method was named an adaptive multi-branch decision tree (CMDT). 11 benchmark data sets with different scales were chosen from UCI Machine Learning Repository and coupled with 12 classifiers to evaluate the CMDT algorithm. The results showed that CMDT can be more reliable than the twelve comparative approaches, especially for imbalanced datasets. We also discussed the performance metrics and the weighted decision-making table in unbalanced data sets. The CMDT algorithm can be found here: https://github.com/chenyuan0510/CMDT.

State of health (SOH) estimation and remaining useful life (RUL) prediction can ensure reliable and safe system operation and reduce unnecessary maintenance costs. In this paper, to improve the accuracy and reliability of SOH estimation and RUL prediction, a novel method based on second-order central difference particle filter (SCDPF) is proposed. By optimizing the importance probability density function, the particle degeneracy phenomenon of particle filter (PF) can be solved. Experiments from the National Aeronautics and Space Administration (NASA) and the Center for Advanced Life Cycle Engineering (CALCE) of the University of Maryland are conducted to demonstrate the effectiveness and satisfactory performance of the proposed SCDPF approach. The maximum error and the root mean square error (RMSE) of the SCDPF fitting approach are quite small, the minimum values of those are 0.006102 Ah and 0.001599, which are lower than those of the unscented particle filter (UPF) and particle filter (PF). The average RUL errors and average PDF width of SCDPF method are also smaller, which validates the accuracy and stability of the proposed method.

This paper investigates a robust prescribed performance control scheme for motion control system with unknown dead-zone input and impactive disturbance. The unknown dead-zone input is divided into a linear part and a disturb part firstly. A smooth dead-zone inverse is constructed to compensate the influence. Subsequently, a novel error transform function is proposed for the limitation that the former prescribed performance control scheme cannot tolerate the impact disturbance. A robust control scheme with prescribed convergence rate is derived with the help of back-stepping technique and the dead-zone parameters is estimated by a designed adaptive law. Lyapunov function is employed to improve the stability of the system. Moreover, a fast finite-time-converge sliding mode differentiator is introduced to estimate the unavailable states of the motion control system. Finally, a two-link robot arm with unknown dead-zone input and high impact is employed to demonstrate the robustness and effectiveness of the control scheme.

To avoid the inconvenience of retyping a user's ID and password, most mobile apps now provide the automatic login feature for a better user experience. To this end, auto-login credential is stored locally on the smartphone. However, such sensitive credential can be stolen by attackers and placed into their smartphones via the well-known credential-clone attack. Then, attackers can imperceptibly log into the victim's account, which causes more devastating and covert losses than merely intercepting the user's password. In this article, we propose a generalized Android credential-clone attack, called data-clone attack. By exploiting the new-found vulnerabilities of original equipment manufacturer (OEM)-made phone clone apps, we design an identity theft method that overcomes the problem of incomplete credential extraction and eliminates the requirement of root authority. To evade the consistency check of device-specific attributes in apps, we design two environment customization methods for app-level and operating system (OS)-level, respectively. Especially, we develop a transparent Android OS customization solution, named CloneDroid, which simulates 101 special attributes of Android OS. We implement a prototype of CloneDroid and the experimental results show that 172 out of 175 most-downloaded apps' accounts can be jeopardized, such as Facebook and WeChat. Moreover, our study has identified 18 confirmed zero-day vulnerabilities. Our findings paint a cautionary tale for the security community that billions of accounts are potentially exposed to Android OS customization-assisted data-clone attacks.

In this paper, the problem of direction-of-arrival (DOA) estimation for a uniform linear array with single-snapshot observations is addressed. Two non-parametric DOA estimators are developed, which can be applied in any azimuth range with one snapshot. Their main idea is iteratively updating the DOA estimates using the weighted least squares and covariance matrix. Two criteria for implementing the covariance matrix are devised, which guarantee high resolution of the proposed methods. The simulation results are included to demonstrate the superiority of our algorithms over several conventional DOA methods in terms of both estimation performance and computational complexity.

A fast and accurate non-iterative direction-of-arrival (DOA) estimation algorithm for multiple targets in additive white Gaussian noise is devised in this paper. The proposed estimator makes use of the two highest magnitudes discrete Fourier transform (DFT) coefficients of the input data and two of their associated neighboring bins, resulting in a deterministic complexity of O(N(1 + log(N))) with N being the number of sensors. The bias and mean squares error of the DOA estimates are analyzed. The simulation results are presented to validate the correctness of theoretical derivation and demonstrate the superiority of the devised estimator over several conventional DOA estimators.

In this paper, the practical performance of space-frequency block code (SFBC)-based multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) system in the presence of nonlinear power amplifier with imperfect channel state information (CSI) over two-wave with diffuse power (TWDP) channel is analyzed. In the performance analysis, we first introduce the equivalent single-input single-output (SISO) scalar model with nonlinear power amplifier and then the SISO model is used to construct a model for the nonlinear MIMO-OFDM system. Considering the channel estimation error at the receiver, we study the effects of MMSE channel estimation error on the channel capacity. What is more, based on inter modulation product (IMP) analysis, we study the BER performance of nonlinear SFBC MIMO-OFDM systems, considering M-ary phase shift keying (MPSK) and M-ary quadrature amplitude modulation (MQAM). And the effects of channel estimation error on the BER performance are studied. Finally, this paper also derives the closed-form expressions of the system optimal operating point. The numerical results and comparisons are provided for several forms of SFBC MIMO-OFDM. From these analysis results, it is found that the analytical interpretation of the observed behavior in the simulation can be advanced, and the actual analysis of the performance optimization of the nonlinear MIMO-OFDM system is also provided.

Automatic diagnosis and classification of schizophrenia based on functional magnetic resonance imaging (fMRI) data have attracted increasing attention in recent years. Most previous studies abstracted highly compressed functional features from the view of brain science and fed them into shallow classifiers for this purpose. However, their classification performance in practical applications is unstable and unsatisfactory. As an acute psychotic disorder, schizophrenia shows functional complexity in fMRI data. Therefore, additional features and deep classification methods are needed to improve classification performance. In this study, we propose a multiple feature image capsule network ensemble approach for schizophrenia classification. The proposed approach proceeds in three steps: 1) extracting multiple image features from the perspective of linear sparse representation, nonlinear multiple kernel representation, and function connection of brain areas respectively; 2) feeding these image features into three specially designed independent capsule networks for classification; 3) obtaining the final results by fusing the outputs of these three deep capsule network using a ensemble approach. To further improve the classification performance, we design a optimization model of maximizing the square of correlation coefficients and propose a weighted ensemble technology based on this model, which is mathematically proved to be solved as a eigenvalue decomposition problem in certain case. Finally, the proposed approach is implemented and evaluated on the schizophrenia fMRI dataset from COBRE, UCLA and WUSTL. From the experimental results, we conclude that the proposed method outperforms some current methods and further improves the accuracy of schizophrenia classification.

Binocular intersection is practically indispensable for stereo visual systems that extract 3D information from images of two distinct cameras. This paper proposes a multi-UAV binocular intersection approach where the two distinct cameras are fixed separately on multiple flying unmanned aerial vehicles (UAVs), rather than physically connected with each other as in the pre-existing works. Eventually, both modeling and algorithms are developed for the multi-UAV binocular intersection with extremely limited communication. As the formulated model states, once one flying vehicle detects the target and shares related information only in a one-shot manner, the other vehicles steer their pan-tilt units (PTUs) to reconstruct a spatial binocular intersection automatically. The binocular intersection algorithm is further improved from two aspects to maximize the probability of target capture. On one hand, the search scope is narrowed by developing an iterative algorithm with priori terrain height and variance. On the other hand, an approaching navigation law is designed to reduce view occlusion caused by terrain. Extensive simulations with different target speeds and flight altitudes are carried out to evaluate the effectiveness of the proposed method. The results show that the probability of binocular intersection is over 90% and is subject to little fluctuation against episodic parameter changes. Furthermore, the developed algorithms are verified localization-free and time-saving as well, and hereafter are potentially practicable for micro aerial swarm applications with extremely limited onboard computation and communication.

The micro-level customer purchase intent in promotions is crucial for the overall purchase conversion rate of promotions. In the context of joint promotions on multiple online channels, customers can access and compare prices and services by navigating between channels to aid their purchase decision. The interactions between customer and promotion channels offer another angle to predict their purchase intent during promotions. In this paper, we propose a feature-combined deep learning framework, in which a full-connected long short-term networks (FC-LSTM) is used for modeling the interactions between customers and promotion channels, as well as the nonlinear sequence correlations and cumulative effects between customer's browsing behavior. To improve the performance of the prediction, the framework incorporates other features of customer profile including purchase history and demographics, integrating them into an end-to-end framework. We apply our method in a real prediction task for online multichannel promotion for concert tickets. Extensive experiments show that the proposed approach exhibits overall good performance compared with state-of-the-art methods on standard metrics such as precision, recall, f-measure, area under curve (AUC), and lift.

Synthesizing sketches from facial photos is of great significance to digital entertainment. Along with higher demands on sketch quality in a complex environment, however, it has been an urgent issue on how to synthesize realistic sketches with the limited training data. The existing face sketch methods pay less attention to the insufficient problem of the training data, leading to the synthesized sketches with some noise or without some identity-specific information in real-world applications. Target on providing sufficient photo-sketch pairs to meet the demand of users in digital entertainment, we present a cross-domain face sketch synthesis framework in this paper. In the photo-sketch mixed domain, we leverage the generative adversarial network to construct a cross-domain mapping function and generate identity-preserving face sketches as the hidden training data. Combined it with the insufficient original training data, we provide sufficient training data to recover the underlying structures and learn the cross-domain transfer of the high-level qualitative knowledge from the photo domain to the sketch domain by the latent low-rank representation. The qualitative and quantitative evaluations on the public facial photo-sketch database demonstrate that the proposed cross-domain face sketch synthesis method can solve the insufficient problem of the training data successfully. And it outperforms other state-of-the-art works and generates more vivid and cleaner facial sketches.

In this paper, we report the narrow-band tunable add-drop filter (ADF) based on the hollow bottlelike microresonator with two taper fibers corresponding to the bus and drop waveguides. We theoretically and experimentally study the relationship between the drop efficiency and the coupling losses of the bus waveguide resonator and the add-drop waveguide resonator. The tuning range of the ADF of 270 GHz (~2.2 nm) and the drop efficiency of approximately 76% have been realized. By such ultranarrow-band filter based on high Q microresonators, a laser beam with specified frequency is filtrated from the background laser with frequency interval of 86 MHz. This tunable ultranarrow-band ADF has potential for applications in the quantum optomechanical experiments and narrow-band quantum photon sources by filtrating the weak signals from pump and noise backgrounds.

In this paper, a compact filtering rat-race coupler implemented in low-temperature co-fired ceramic (LTCC) technology is presented. In order to reduce the dimensions, the schematic and the coupling topology of a planar filtering rat-race coupler are first studied and then simplified to fit the 3-D multilayer LTCC structure. An eight-line spatially symmetrical coupled structure is, therefore, proposed to obtain the required coupling topology between resonators. Thus, the filtering rat-race coupler can be constructed only by using four resonators, which is only half that of the planar one. Consequently, the size could be reduced substantially. For validation, a design example centered at 3.5 GHz is simulated, fabricated, and measured. All the measured results are in good agreement with the full-wave simulation results. Besides, the size of the circuit is only 3.8×4×2.1 mm3, which demonstrates that by employing the eight-line spatiallysymmetrical coupled structure, the proposed rat-race coupler obtains good performance and compact size.

Patch antennas with multiple feeds are proposed in this paper. The antennas consist of a traditional square patch and feeding probes that are set in a line orthogonal to the polarization direction. In addition to radiating EM energy just like the conventional antennas, the proposed multi-port antennas can also realize on-antenna power combining without power combiner. Thanks to this, high-line losses and large system size caused by extra power combiner can be avoided. Moreover, since the location and the number of ports have a limited effect on the antenna performances, flexible feeding strategies can be chosen in different applications. For demonstration, four patch antennas with a different number of ports (one port, two ports, three ports, and four ports) are designed at the center frequency of 1.5 GHz. Compared with the traditional one-port patch antenna, the proposed three multi-port patch antennas can achieve nearly the same antenna performance, including bandwidth, polarization, efficiency, and radiation pattern. Besides, one-port antenna and two-port antenna integrated with a two-way power amplifier are designed and measured, respectively. Then, they are compared to further prove the proposed power combining method in multi-port antennas.

The problem of data reconstruction with partly sampled elements under a tensor structure, which is referred to as tensor completion, is addressed in this paper. The properties of the rank-1 tensor train decomposition and the tensor Kronecker decomposition are introduced at first, and then the tensor Kronecker rank as well as Kronecker rank-1 tensor train decomposition are defined. The general tensor completion idea is presented following the criterion of minimizing the number of Kronecker rank-1 tensors, which is relaxed to the thresholding problem and the solution is derived. Furthermore, the number of Kronecker rank-1 tensors that the proposed algorithm can retrieve and its complexity order are analyzed. Computer simulations are carried out on real visual data sets and demonstrate that our method yields a superior performance over the state-of-the-art approaches in terms of recovery accuracy and/or computational complexity.

This paper proposes a co-simulation platform using field-programmable gate array (FPGA) and real-time digital simulator (RTDS) for the simulation of large power systems. It combines the advantages of high computational power from FPGA and better modelling flexibility from RTDS together. The FPGA therefore acts as an efficient and economical extension to RTDS especially when simulating large ac systems. One of the significant advantages of the proposed co-simulator is that it avoids the potential interface error existing in the conventional approach of interfacing transient stability program with electromagnetic transient programs. Two key aspects of the proposed co-simulator are discussed: 1) the interface design between FPGA and RTDS and 2) the hardware implementation and expandability of the platform. Two case studies are presented to verify the simulation accuracy and capability of the proposed co-simulator. The first case simulates a two-area four-machine power system with one area simulated in FPGA and the other area in RTDS. Comparisons are made with the case where the complete system is simulated in RTDS. The second case simulates a system of 141 buses in FPGA to demonstrate the simulator's capability in simulating large power systems.

The authors inadvertently left out an author to the above paper [1]. His name is Yiming Lei, and he is the corresponding author to this paper.

Carbon nanotube (CNT)-based multibeam X-ray tubes provide an array of individually controllable X-ray focal spots. The CNT tube allows for flexible placement and distribution of X-ray focal spots in a system. Using a CNT tube, a computed tomography (CT) system with a noncircular geometry and a nonrotating gantry can be created. The noncircular CT geometry can be optimized around a specific imaging problem, utilizing the flexibility of CNT multibeam X-ray tubes to achieve the optimal focal spot distribution for the design constraints of the problem. Iterative reconstruction algorithms provide flexible CT reconstruction to accommodate the noncircular geometry. Compressed sensing-based iterative reconstruction algorithms apply a sparsity constraint to the reconstructed images that can partially account for missing angular coverage due to the noncircular geometry. In this paper, we present a laboratory prototype CT system that uses CNT multibeam X-ray tubes; a rectangular, nonrotating imaging geometry; and an accelerated compressed sensing-based iterative reconstruction algorithm. We apply a total variation minimization as our sparsity constraint. We present the advanced CNT multibeam tubes and show the stability and flexibility of these new tubes. We also present the unique imaging geometry and discuss the design constraints that influenced the specific system design. The reconstruction method is presented along with an overview of the acceleration of the algorithm to near real-time reconstruction. We demonstrate that the prototype reconstructed images have image quality comparable with a conventional CT system. The prototype is optimized for airport checkpoint baggage screening, but the concepts developed may apply to other application-specific CT imaging systems.