Search

Object detection in aerial images is vital for autonomous guidance, navigation and control, and situational awareness. However, there are still many challenges facing researchers in this filed, including the target scales, the perspectives in taking pictures, and the highly complex background. The present paper introduces a robust object detector which is optimized for handling with multi-scale objects and the overhead capturing perspective object instances in aerial images. Firstly, in the feature extraction stage, an effective multi-scale detector (MSD) is designed to search for objects with different scales in feature maps. After that, when detecting a small target from a cluttered background, both the shallow and deep layer features are densely connected by the deconvolution after tackling the issues of low dimensionality in deep layers and inadequate representation of small objects. In the experiments part, we analyze the impacts of the above mentioned components on the model and make a comparison between the method at issue and other state-of-the-art approaches on two publicly-available datasets captured by satellites and high-altitude UAVs. The results show that the proposed method, which is applicable to a wider range of aerial images, is more effective and robust.

The output power of a wireless power transfer (WPT) system depends highly on the load and the coupling conditions, which are variable in practical applications generally. In order to attain stable output power with high transfer efficiency under various coupling condition and load, this paper proposes a series-series compensated WPT system based on parity-time (PT) symmetry with front-end DC-DC converter and a novel primary-side-only control strategy. The strategy comprises the negative resistance control of PT-symmetric circuit, the online load identification approach by PT-symmetry, and the power closed-loop control method. The advantages are that the mutual inductance information is not needed, and dual-side wireless communication as well as the secondary-side control circuit is avoided, which compresses the volume of the secondary side, simplifies the control algorithm, and improves the robustness of the system. With the proposed primary-side-only control strategy, the output power is coupling-independent and can automatically stabilize at specified values over reasonable variations both in coupling coefficient and load. The experimental results obtained from a prototype are included. They confirm the proposed control strategy and indicate that system can stably output 200 W and 400 W with the maximum error 4.10% and 3.40% respectively when the coupling coefficient and loads vary, and achieves high overall efficiency at 91.9%.

As an essential and challenging technology of fault prediction and health management(PHM), fault prediction technology has been a research focus in the field of fault diagnosis. However, the current model-based fault prediction technology and data-driven fault prediction technology have some limitations, and it is difficult to effectively apply them in practice. Therefore, this paper combines the advantages of two kinds of fault prediction technology, sets the fault distribution function as the membership function of the adaptive fuzzy neural network based on the full analysis of the fault mechanism. The use of the fault distribution function to highly generalize the law of fault occurrence, and the strong self-learning ability of the neural network can effectively tap the potential fault information of the fault data, thereby using the fault distribution function to fit the fault data, and forming a set of membership functions by presetting a variety of membership functions, so as to expand the applicability of the proposed model in fault prediction. The experimental results show that the fault prediction model proposed in this paper has the advantages of high prediction accuracy, fast convergence speed and good applicability.

In this paper, the development of two high power 220 GHz frequency triplers is proposed. The GaAs Schottky diodes with six nodes are applied to realize high efficiency 220 GHz tripler, while the application of GaN Schottky diodes with eight nodes is another attempt to improve power handling of the 220 GHz tripler. To reduce thermal effect of high power multipliers, the AlN substrates with high thermal conductivity are applied to provide better heat dissipation at the diode areas. A combination of electrical and thermal model of the Schottky diodes is established while the optimization of 220 GHz triplers are realized with 3D electromagnetic (EM) simulation and harmonic balanced simulation. Good agreement is achieved between the simulated results based on electro-thermal model and measured performances of the triplers. At room temperature, peak efficiency of the tripler based on GaAs Schottky diodes is 17.8%, while the maximum output of the tripler is 38.2 mW with 300 mW input power. As for the 220 GHz GaN Schottky diode tripler, measured results show that the maximum power handling is beyond 400 mW. The peak efficiency and maximum output are 4.7% and 18.4 mW, respectively. The proposed methods of developing high power multipliers can be applied in higher frequency band in the future.

Increasing near-surface temperature over the Qinghai-Tibet Plateau (QTP) has led to permafrost degradation and increasing active layer thickness (ALT). In this study, the ALT was estimated based on ground subsidence monitored by multitemporal interferometric synthetic aperture radar (MT-InSAR) and temporal-spatial multilayer soil moisture data. For the ground subsidence monitoring, a modified Stefan piecewise elevation change model based on air temperature data was integrated into a new small baseline subset (NSBAS) chain. A total of 33 scenes of Sentinel-1 data (S-1) were collected over one year to build the MT-InSAR analysis network. Moreover, both soil moisture active/passive (SMAP) L4 surface and root zone soil moisture data and ERA-Interim reanalysis data were used to build an ALT retrieval model. In particular, the global-scaled soil moisture data (SMAP and ERA-Interim) fraction was separated based on the Sentinel-1 amplitude-based land cover classification results and in situ soil moisture data. A typical ALT estimation method based on the point scale groundwater information was also performed to evaluate the performance of the proposed method. Based on the validation of the ground-based ALT observations, the proposed method outperformed the traditional point scale groundwater information-based method, with a correlation coefficient of 0.67, RMSE of 0.70 and ubRMSE of 0.51, respectively. The ERA-Interim-based estimation results were underestimated due to the overestimation of the ERA-Interim soil moisture data. Obvious differences were observed between the ALT of the alpine meadow areas and alpine desert areas. Our results demonstrate that the combination of temporal-spatial multilayer soil moisture data and the MT-InSAR method with S-1 images is a promising approach for the large-scale characterization of ALT.

This study proposes a three-dimensional (3-D) absolute attitude reconstruction algorithm of a rigid body target based on multi-station high-resolution range profile (HRRP) sequences. The existing 3-D reconstruction algorithms rely on the target motion form and the reconstructed attitude is inaccurate due to the mismatch of the micro-motion model. In this study, the coordinates of the scattering centers are first calculated from the HRRP sequences of each radar using the factorization method. Then, the multi-station data are associated with the reconstructed scattering center coordinates, and the absolute coordinates is estimated by the relationship between the associated radar line-of-sights (LOSs) and the absolute LOSs. The 3-D reconstruction algorithms for the stable scattering centers and the sliding scattering centers are studied. Numerical simulations verify that the proposed algorithm can reconstruct the 3-D absolute attitude of a rigid body target and is not affected by the mismatch of the target micro-motion model.

As the system becomes more and more complex, we are usually in the state of indeterminacy. In the real world, the states of uncertainty and randomness are the two most common types of indeterminacy. An uncertain random graph is applied to describe a graph model with uncertainty and randomness simultaneously. This paper mainly focuses on the connectivity of two vertices in an uncertain random graph. Firstly, a local connectivity index is proposed to unveil the chance measure that two special vertices are connected in an uncertain random graph. Furthermore, a method for calculating the local connectivity index is formulated. In addition, some simplified forms of the method are developed, and an algorithm is designed to obtain the local connectivity index. Finally, the information relevant to the relationship between the local connectivity index and the connectivity index is discussed.

The ocean environment is complex and changeable because of all kinds of noise interferences, such as salt cliffs, ships around and other electromagnetic interferences, so the measurement information is prone to be lost. It is difficult to describe the complex noise and the acquisition probability of measurement information. In this paper, a continuous discrete variational Bayesian filter (CD VBF) is proposed to solve the problems of the heavy tailed noises and measurement random loss for state estimation. The variational Bayesian (VB) approach can effectively estimate the state vector, scale matrices, degree of freedom (DOF) parameters, Bernoulli random variables and the acquisition probability of measurement information. The performances of the proposed algorithm and traditional algorithms are tested in simulations and underwater experiments. The simulation results illustrate that the CD VBF has better localization accuracy and robustness. The experimental results demonstrate that the localization accuracy is improved by CD VBF and an optimal solution of iteration number is acquired.

Directional modulation (DM) can transmit information to the desired direction or directions with known constellation mappings, but with scrambled ones in other directions. However, one problem of the design is that when desired receivers and eavesdroppers are in the same transmission direction, their beam responses cannot be distinguished, as steering vectors for these locations are the same. To solve the problem, positional modulation (PM) is introduced where a given modulation pattern can only be received at certain desired positions. In this work, the multi-path effect is exploited for positional modulation with the aid of metasurface acting as a low-cost flexible reflecting surface. Design examples are provided to show the effectiveness of the proposed design.

The global positioning system real-time kinematic (GPS-RTK) instrument is a commonly used instrument to obtain GPS coordinates in cadastral surveying, but its use is limited to unshielded environments. This study proposes a new method based on a visual simultaneous localization and mapping (SLAM) algorithm that uses a stereo-camera to replace traditional methods. The benefits of our method are a higher measurement efficiency, fewer required professionals, and fewer restrictions on the operating environment compared with the traditional RTK and total station method. First, we propose a robust non-planar camera calibration algorithm that uses three non-planar chessboards to obtain intrinsic parameters of a stereo-camera to improve the accuracy of the measurement. Second, we propose a detailed algorithm that uses a stereo-camera and known GPS coordinates to obtain unknown GPS coordinates. Using experiments, we demonstrated the effectiveness and stability of our method to provide accurate GPS coordinates.

This paper presents and analyzes an AC-DC power converter structure, which is comprised of a Power Factor Correction (PFC) module and a LLC resonant DC-DC converter module. This converter only uses two switches, and requires three less diodes and one less switch compared to popular LLC resonant converter solutions. Compared to its conventional counterpart, the rectifier of interest has high energy efficiency while a smaller size, owing to the soft-switching in the LLC resonant converter. Detailed theoretical analyses are conducted in this study, followed by software simulation and hardware experimentation, which demonstrate that the single stage double-switched (DS)-LLC rectifier is able to realize unity power factor and a wide output range, indicating its effectiveness and applicability.

High frequency and high efficiency operation is one of the premier interests in the signal and energy conversion applications. The wide bandgap GaN based devices possess superior properties and have demonstrated exceeding performance than Si or GaAs devices. In order to further exploit the potential of GaN electronics, monolithic power integration is proposed. Firstly, this paper discusses the structure and properties of GaN power devices to explain the choice of lateral integration in the view of GaN power ICs. Then the state-of-the-art performance of GaN power integration in two major application areas is reviewed, which are the microwave power amplification and DC-DC power conversion. The GaN power integration technologies in MMIC platforms are summarized in terms of the gate length, operation frequency and power added efficiency of ICs. On the other hand, the smart GaN power IC platforms have boosted the development of DC-DC power converters. Demonstrations of high frequency (>1 MHz) and high efficiency (>95 %) converters with various kinds of integration technology and topology are reviewed. Lastly novel integration schemes and methods are introduced to stimulate new thoughts on GaN power integration road.

The WPT system for logistic robots has the problems of large fluctuation of charging power and low efficiency due to the change of its position. Based on the principle of parity time symmetry, an omnidirectional and efficient wireless power transfer system for logistic robots is proposed. Firstly, the coupling coefficient of two planar spiral coils in different relative positions in space are studied. Then, a time-varying coupled-mode model of the WPT system based on the offset angle of the receiver coil is established, and its working principle and characteristics are analyzed. As long as the coupling coefficient satisfies certain conditions, the output power and efficiency of the system are always keeping constant. Finally, the proposed model is experimentally verified. The measured results show that the WPT system proposed can realize stable 150W power transfer with constant transmitting efficiency of more than 90% for a logistic robot within a confined three-dimensional space around the charging station, which are provided to further verify the correctness of theoretical analysis.

In this study, the problem of transmitted waveform design for a multiple-input-multiple-output (MIMO) radar in a heterogeneous clutter environment is considered. Under a Bayesian framework, the waveform design problem is used to maximize the generalized likelihood ratio for a given false alarm probability. In addition, the proposed problem is nonconvex under the constant modulus constraint. Hence, the alternating direction method of multipliers-based and the Broyden-Fletcher-Goldfarb-Shanno method are adopted to solve this nonconvex optimization problem. Finally, the clutter suppression performance of the proposed algorithms is evaluated in real-world data by numerical simulations. The simulation results show that the proposed methods have better performance on clutter suppression and target detection.

Binary butterfly optimization approach (bBOA) is a recent high performing feature selection algorithm presented in 2018 which is based on the food foraging behavior of butterflies. This paper tries to improve the structure of the bBOA to enhance its classification accuracy, dimension reduction and reliability in feature selection task for who are interested in the fields of data mining and pattern recognition. The new initialization strategy and differential evolution strategy are applied to reduce the randomness of bBOA's initialization and local search process. Then, a new parameter is added to make the bBOA's transfer function more adaptive to the change of exploration and exploitation. Besides, evolution population dynamics (EPD) mechanism is employed as an extension of bBOA. The new method called optimization and extension of binary butterfly optimization approaches (OEbBOA) is tested with the K nearest neighbor classier in which twenty UCI datasets and seven recent algorithms are utilized to assess the performance of the OEbBOA algorithm. The experimental results and nonparametric Wilcoxons rank sum test confirm the efficiency of the proposed OEbBOA in maximizing classification accuracy while minimizing the number of features selected.

The measurement of semantic similarity between concepts is an important research topic in natural language processing. In the past, several approaches for measuring the semantic similarity between concepts have been proposed based on WordNet or Wikipedia. However, improvements in the measurement accuracy of most methods have led to a dramatic increase in time complexity, and the existing methods do not effectively integrate WordNet and Wikipedia. In this paper, we focus on designing an efficient semantic similarity method based on WordNet and Wikipedia. To improve the accuracy of WordNet edge-based measures, we propose an edge weight model for combining edge and density information, which assigns a weight to each edge adaptively based on the number of direct hyponyms of the subsumer. Second, to improve the computational efficiencies of the existing Wikipedia link vector-based measures, we propose a new Wikipedia link feature-based semantic similarity method that converts Wikipedia links into semantic knowledge and replaces the TF-IDF statistical weight model in the existing measures. In addition, we propose two new word disambiguation strategies to further improve the accuracy of Wikipedia link-based measures. Finally, to fully exploit the advantages of WordNet and Wikipedia, we propose two new aggregation schemas for combining WordNet “is-a” semantics and Wikipedia link semantics to replace the current aggregation schemas that combine WordNet “is-a” semantics with category semantics in Wikipedia. The experimental results show that our aggregation models are outstanding in terms of accuracy, efficiency and word coverage compared to state-of-the-art similarity measures.

This paper deals with a hybrid non-isolated active quasi-switched dc-dc converter with a high-boost voltage gain, which is applicable for the high step-up low power applications. By using fewer number of components in circuit topology, the proposed converter can provide higher-gain voltage with a small duty cycle, which can reduce the voltage stress and conduction loss on power switches. In addition, it draws continuous input current, has lower diode voltage stress and lower passive component voltage ratings. The operating principles and key waveforms in continuous conduction mode and discontinuous conduction mode are presented. Small-signal dynamic analysis, parameter design guideline, power loss calculation, and characteristics comparison with other non-isolated converters have been completed. Finally, a 250W hardware prototype is constructed and the experimental results are presented to verify the feasibility of the proposed converter.

A modular multilevel converter (MMC) is an emerging converter technology widely used in wind power applications using high-voltage direct current (HVDC) transmission, and it has been largely investigated in medium-voltage solar harvesting as well as electric motor drives. Different from studies dealing with active circulating current control loops, this paper focuses on impedance modeling and harmonic stability studies for MMCs with passive circulating current filters (PCCFs). A power stage circuit including a PCCF is transformed and remodeled to obtain small-signal formulations. This transformation leads to additional high-order matrix computing complexity due to the added impedance subnetwork matrix from the PCCF as well as relevant crossed-frequency impacts. To simplify the computational burden, the proposed model aims to solve one arm equation instead of all six MMC arm equations. To achieve this result, additional challenges occur when the MMC is connected to a renewable energy source instead of the grid voltage, where low-level zero-sequence components might exist in the neutral point of three phases, especially in the case of no integrated advanced modular voltage balancing control, which prevents computational simplification. To overcome this challenge, further impedance matrix adjustments are conducted in this paper to theoretically suppress the impact of zero-sequence harmonics when obtaining small-signal impedance, taking into account the frequency coupling effect. Finally, simulations are carried out to validate the developed impedance model under different operating scenarios. Harmonic stability case studies of MMCs with PCCFs connected to renewable energy current sources are also presented, where frequency-domain stability analyses based on Bode diagrams are compared to time-domain simulation waveforms and FFTs, which validate the effectiveness of the proposed impedance model.

Positional modulation (PM) has been introduced recently where a given modulation pattern can only be received at certain desired positions. To achieve it, the multi-path effect is exploited for positional modulation with the aid of metasurface acting as a low-cost flexible reflecting surface. In this paper, sparse antenna array based positional modulation design is proposed for the first time; to reduce the implementation complexity of the metasurface, the number of active units is also minimised for a given PM design requirement. Design examples are provided to show the effectiveness of the proposed design.

As a key technology of the 5G communication system, device-to-device (D2D) communications have drawn significant research interests. However, the advantage of D2D communications may be limited when D2D users are far away from each other or the communication environments are harsh. In order to extend the range of D2D communications, a promising way is to use relaying technique to assist the communications. In this paper, we use a two-way amplify-and-forward (AF) relay to assist the underlay D2D communications and investigate the power control problem. Specifically, we formulate the power control problem as the optimization of the performance of the D2D link while fulfilling the quality-of-service (QoS) requirement of the cellular link. Two optimization objectives are considered for the D2D link, i.e., maximization of the minimum SINR at D2D users and maximization of the sum-rate of D2D users. For the first optimization objective, we can compare the possible optimal solutions for the three boundary conditions and obtain the optimal solutions. For the second optimization objective, we can convert the objective function into a concave function based on difference of convex (D. C.) structure, and use an iterative algorithm to solve the optimization problem. The simulation results show that the proposed power control scheme can improve the performance of the D2D link.

The penetration of renewable energy generations, e.g., wind power, not only introduces the randomness and fluctuations into power system operations but also increases the possibilities of cybersecurity issues. Among them, false data injection attack (FDIA) can access and falsify the readings of smart meters, which would impede the functionalities of power systems. In this paper, we first set up an evaluation model to identify the set of high-risk lines by investigating the relationship between FDIA and wind power uncertainty. Then, for a power system with a high wind penetration, a tri-level preventive dispatch strategy is proposed to ensure the system security even under the worst-case of FDIA. It is demonstrated that the impacts of FDIA can cause more serious security issues as the wind penetration level increases. The effectiveness of the proposed tri-level preventive dispatch strategy in mitigating the FDIA caused overloading risk is validated using the IEEE 118-bus system.

With the wide deployment of advancing metering infrastructure (AMI) in power distribution systems, the quantity of power consumers' electricity data is increasing rapidly and the data also become more and more accurate. To make full use of these power consumers' AMI data, a data-driven abnormity assessment algorithm for low-voltage power consumers is proposed based on the CRITIC (CRiteria Importance Though Intercrieria Correlation) method and the improved radar chart method. First, the indexes that characterize the consumer's abnormal features of power consumption and supplies are extracted from the original AMI data. Then, the abnormity assessment algorithm is used to determine power consumers' abnormal features of power consumption and supplies by using the extracted indexes, in which the weights of indexes are determined by the CRITIC method and the assessment value of abnormal features is determined by the improved radar chart method. Next, the abnormity assessment algorithm is used again to assess power consumers' power consumption and supplies abnormities. Finally, the effectiveness of proposed algorithm is demonstrated in case studies by employing AMI data collected from power utilities in Zhejiang Province, China, and the results show that the algorithm can be used in actual applications.

Global navigation satellite system-reflectometry (GNSS-R) has attracted much attention in remote sensing, such as sea wind speed detection, soil moisture, and sea ice. In order to verify the algorithm of GNSS-R, the simulation of GNSS-R plays an important role in this research field. Delay waveform (DW) and delay-Doppler map (DDM) are the main observables in GNSS-R simulation, and are usually effects by the complex ocean environment, such as wind and swell. Besides the effects of wind, this article focuses on analyzing the influence of swell on the simulation of GNSS-R. In this article, a complex model of DW and DDM is presented with effects of swell. Swell modeled here by narrow-band Gaussian spectrum, and the influence of swell on the sea surface scattering coefficient, which can be calculated using Kirchhoff approximation-geometric optics. Observables obtained from the simulation in the space-borne scene are analyzed. The simulation results show that swell has a stable and significant impact on the simulation of GNSS-R observables. Compared with models under no consideration of swell, the model with effects of swell has higher accuracy.

Metasurface based positional modulation design has been introduced recently, where a given modulation pattern can only be received at certain desired positions by the proposed method. However, the magnitude of weight coefficient of each element on metasurface varies from one other, representing an attenuation of the amplitude of incoming signal in different degrees. In this paper, a constant reflection attenuation constraint for incoming signals is proposed for the first time, and the proposed method can be extended with a post-processing process to the ideal case where there is no reflection attenuation.

People pay greater attention to air quality which is closely related to their health, especially in developing countries. Air quality is a part of the Chinese weather forecast, and the government has developed air quality monitoring systems and built high-quality monitoring stations (HQMS). With the data from HQMS, many companies and research institutes demonstrate an accurate air pollution map on the Internet, which is valuable for many issues related to air quality, including exposure modeling and urban planning. Due to the high equipment and operating costs, the distribution of HQMS is too uneven and sparse to achieve a high-resolution air pollution map. Thus, people try to deploy a large number of high precision sensors in and around the city for detecting air pollution. However, these sensors require frequent calibrations with the HQMS to maintain data accuracy. On the other side, to reduce the cost of sensor deployment, people begin to use mobile sensors instead of fixed sensors, which make sensor route planning a very important issue. Nevertheless, existing works on the route planning of mobile sensors largely focus on data reconstruction, which either ignores calibration or views it as an independent problem. In this paper, we propose a novel scheme to improve the accuracy of data reconstruction, which jointly considers sensor calibration and data reconstruction in route planning for mobile sensors. We formulate a novel sensor route planning problem (SRPP) that aims to maximize the mutual information and guarantee the accuracy of measurements through the sensor calibration. We also propose a heuristic algorithm to solve the SRPP, which supports calibration between mobile sensors and HQMS in route planning. The extensive simulation results well justify the effectiveness of our approach that can reduce 83% root mean square error on average compared with the traditional approach.

Girth is an important structural property of low-density parity-check (LDPC) codes. Motivated by the works on the girth of Tanner (3, 5), (3, 7), (3, 11), and (5, 7) quasi-cyclic (QC) LDPC codes, we, in this paper, study the girth of Tanner (3, 13) QC-LDPC codes of length $13p$ for $p$ being a prime of the form $(39m+1)$ . First, the cycle structure of Tanner (3, 13) QC-LDPC codes is analyzed, and the cycles of length lesser than 12 are divided into five equivalent classes. Based on each equivalent class, the existence of the cycles is equivalent to the solution of polynomial equations in a 39th unit root in the prime filed $\mathbb {F}_{p}$ . By solving these polynomial equations over $\mathbb {F}_{p}$ and summarizing the resulting candidate prime values, the girth of Tanner (3, 13) QC-LDPC codes is obtained. As an advantage, Tanner (3, 13) QC-LDPC codes have much higher code rates than Tanner (3, 5), (3, 7), (3, 11), and (5, 7) QC-LDPC codes, and provide a promising coding scheme for the data storage systems and optical communications.

Wireless power transfer (WPT) offers a promising solution to power implantable medical devices (IMDs). Due to the variations of mutual inductance and load, the output voltage of the implantable WPT system is easily unstable. To maintain a constant voltage for IMDs’ operation, a novel method to identify the output voltage of the WPT system without any direct measurement on the receiving side is presented in this paper, where only the input voltage and current need to be measured. First, the output voltage identification of the WPT system is applied to the classic series–parallel compensation network. Next, an improved WPT system with S-LCL compensation network is proposed, which has the advantage that the output voltage identification is independent of mutual inductance and load resistance. Moreover, the characteristic analysis of the proposed WPT system is carried out, which proves its performance of high transfer efficiency. Finally, the WPT prototype with the S-LCL compensation network is built and tested. The experimental results are provided to further verify the correctness of theoretical analysis.

Girth plays an important role in the design of low-density parity-check (LDPC) codes. Motivated by the works on the girth of some classes of Tanner quasi-cyclic (QC) LDPC codes, e.g., Tanner (3, 5), (3, 7), (3, 11), and (5, 7) codes, we, in this paper, study the girth of Tanner $(J,L)$ QC-LDPC codes where $J$ and $L$ can be any two positive integers. According to the sufficient and necessary conditions for the existence of cycles of lengths 4, 6, 8, and 10, we propose an algorithm to determine the girth of Tanner $(J,L)$ QC-LDPC codes with finite code lengths. Through the analysis of the obtained girth values, we generalize the laws of the girth distributions of Tanner $(J,L)$ QC-LDPC codes. Furthermore, based on the exponent matrices of Tanner $(J,L)$ QC-LDPC codes with known girths, we employ the column selection method and/or the masking technique to construct binary/nonbinary LDPC codes. The numerical results show that the constructed LDPC codes have good performance under iterative decoding over the additive white Gaussian noise channel.

This paper proposes a novel 220-GHz third-harmonic mixer, which using a waveguide-fin-coplanar waveguide-microstrip line hybrid transmission line as the carrier circuit, based on the balanced structure and the co-directional diode pair. The hybrid transmission line is designed to allow the mixing signals to be input in the correct direction in order to achieve the third-harmonic mixing. Then, high isolation between the radio-frequency (RF) port and the local oscillator (LO) port is performed through the mode orthogonality of the balanced structure. The metal slots on the fin line are used as the matching network for the RF signal to enable the RF signal into the diode pair with low loss. The experimental test results show that the conversion loss of the third-harmonic mixer is better than 18.5 dB in the range of 210-230 GHz and has a tiny fluctuation in the broad range of intermediate frequency (IF).

Multi-carrier-based waveform design for directional modulation (DM) is studied, where simultaneous data transmission over multiple frequencies can be achieved, with given phase distribution at the main lobe and as random as possible over sidelobe regions for each frequency. The design can be implemented efficiently by the inverse discrete Fourier transform (IDFT) structure. However, the problem of multi-carrier design is the high peak-to-average-power ratio (PAPR) of the resultant signals, leading to non-linear distortion when signal peaks pass through saturation regions of a power amplifier. To solve the problem, the $\text {PAPR}\leq \rho ~(\rho \geq 1)$ constraint is considered in the design, and a solution called wideband beam and phase pattern formation by Newton's method (WBPFN) is proposed. The resultant beam patterns, phase patterns, and complementary cumulative distribution function (CCDF) of PAPR are presented to demonstrate the effectiveness of the proposed design.

Continuous annealing production process generally consists of multiple complex processes that are coupled to each other, and each process contains many control variables. It is difficult to establish a precise mechanism model of the production process. The operators mainly set these control variables based on past production experience, which often result in great fluctuations of product quality (even unqualified products) and high energy consumption. This in turn significantly affected production cost and economic benefits of the cold rolling mill. To efficiently handle this problem, an ensemble learning modeling method based on production data is first proposed for this production process, and then, a multiobjective operation optimization model is established to optimize the operation of continuous annealing production process. Finally, an improved multiobjective differential evolution algorithm based on search process memory is developed to solve this model and achieve the optimal setting of control variables. The computational results on both benchmark problems and practical problems illustrate that the proposed algorithm is superior to some powerful multiobjective evolutionary algorithms in the literature and it can effectively achieve good setting of control variables for the continuous annealing production process.

Nonlinear conductivity dielectric is widely used today to solve high energy discharge problems in many fields (such as spacecraft charging and motor insulation). This paper summarized the research about the nonlinear conductivity theory of composite materials. In particular, the key issues about nonlinear conductivity studies of composite materials are proposed. The nonlinear conductivity mechanisms of two typical filler particles (SiC and ZnO) are introduced. In addition, two kinds of nonlinear conductivity theory, including the traditional dielectric conduction theory and the polymer conduction theory, are reviewed. Several mechanisms in each of the conduction theories are analyzed, and also, the advantages and disadvantages of these theories are presented in detail. Thus, the purpose of this paper is to let the researchers know the nonlinear conductivity research status fully. We believe this paper is of great significance for the nonlinear dielectric conductivity researchers.

Recently, WiFi networks have witnessed an unprecedented deployment, and WiFi APs are ubiquitous around the world. On one hand, the WiFi clients have to actively discover new WiFi access points (APs), and try to access to them, even though most of the APs are privately owned, which wastes the precious energy of mobile devices due to excessive listening and scanning operations of WiFi network interface cards. On the other hand, many wireless technologies coexist in overlapping channels (e.g., 2.4 GHz ISM bands). It provides a new opportunity of cross-technology communication (CTC) that utilizes these overlapping frequencies to share information among different technologies. This paper deeply investigates the application of wireless CTC into smart and energy-effective WiFi access with the assistance of low-power radio. Specifically, we design a smart WiFi access system assisted by Zigbee, SmartWAZ, for mobile terminals with both WiFi and Zigbee interfaces. The most prominent feature of this system is that the different WiFi beacon periods are intentionally used to distinguish publicly accessible and privately owned WiFi APs. Then, the folding algorithm is applied by the Zigbee module in clients to detect the specific WiFi beacon period used by public APs, and then the WiFi interface will be waked only when a public WiFi AP is detected. From the system design viewpoint, we investigate the influence of the beacon period of the sender (i.e., WiFi AP) and the judgment threshold of the mobile client on a false positive and false negative, and provide the basic rule to appropriately set those system parameters. Compared to the traditional WiFi access methods, the SmartWAZ significantly saves the energy consumption of continuous WiFi scanning; compared to other low-power radio assisted WiFi access systems, SmartWAZ avoids waking WiFi interface when private WiFi APs appear. The experiment results show that the SmartWAZ averagely consumes only 36% of the energy of traditional WiFi access schemes and achieves reliable detection with an error rate of less than 1%.

Recent research has witnessed the fostered application of machine learning approaches in analyzing the single nucleotide polymorphisms (SNP) data, which has been proved to be implicated in complex human diseases. In the identification of SNPs responsible for complex diseases, most genome-wide association studies always took single SNP into consideration at one time and ignored diverse interactions between SNPs. One of the major problems is the higher number of features and the relatively small number of individuals, which complicates the task and harms the predictive ability of DNA sequences. In this paper, a novel boosting-based ensemble approach was proposed to study these interactions. An importance scoring strategy based on Gini impurity was introduced for feature selection. We evaluated its efficacy on the SNP genotyping data collected by the Southeastern University of China and compared it with naive Bayes, support vector machine, and random forest. The experimental results have shown its validity and effectiveness on SNP interaction identification. In addition, our approach had an obvious advantage of computational time and resources.

The appearance of generative adversarial networks (GAN) provides a new approach and framework for computer vision. Compared with traditional machine learning algorithms, GAN works via adversarial training concept and is more powerful in both feature learning and representation. GAN also exhibits some problems, such as non-convergence, model collapse, and uncontrollability due to high degree of freedom. How to improve the theory of GAN and apply it to computer-vision-related tasks have now attracted much research efforts. In this paper, recently proposed GAN models and their applications in computer vision are systematically reviewed. In particular, we firstly survey the history and development of generative algorithms, the mechanism of GAN, its fundamental network structures, and theoretical analysis of the original GAN. Classical GAN algorithms are then compared comprehensively in terms of the mechanism, visual results of generated samples, and Frechet Inception Distance. These networks are further evaluated from network construction, performance, and applicability aspects by extensive experiments conducted over public datasets. After that, several typical applications of GAN in computer vision, including high-quality samples generation, style transfer, and image translation, are examined. Finally, some existing problems of GAN are summarized and discussed and potential future research topics are forecasted.

An optoelectronic circuit integrated with a photodiode is proposed, which simultaneously achieves 3.5-MΩ transimpedance gain and 18-MHz bandwidth. The design is intended to address the limits on bandwidth introduced by the large parasitic capacitance of the integrated photodiodes in optoelectronic applications. The proposed circuit consists of a transimpedance amplifier (TIA) and a voltage gain stage. The pre-stage TIA with low input impedance provides the main transimpedance gain of the circuit over a wide bandwidth. The post voltage gain stage provides the supplemental voltage gain for the circuit and increases its driving capability. The circuit is fabricated in a 0.5-μm CMOS process with an active area of 501.2 μm × 291.6 μm. When the light input is a 100-ns pulse with 905-nm wavelength and 1.5-μW optical power, the amplitude of output signal is 1.6 V with a rise time of 24.5 ns and a delay time of 21 ns.

In this paper, a novel WR-2.8 band (260-400 GHz) pseudo-elliptic waveguide bandpass filter (BPF) based on a singlet and an extracted pole resonator is proposed. The singlet based on TE301 mode and extracted pole resonant cavities is developed to generate two transmission zeros (TZs) on both sides of the passband in order to achieve high selectivity. The mechanism of TZs generation of both structures is analyzed in this paper, and TZs position is predicted precisely. Furthermore, each TZ position can be controlled independently by changing the dimensions of resonant cavities. The proposed filter fabricated by conventional computer numerical control (CNC) milling technology exhibits an insertion loss (IL) around 0.7dB, a 3-dB fractional bandwidth (FBW) of 9.9% centered at 357GHz and an ultra-high selectivity with a 0.87 30-dB rectangular factor which is all in good agreement with the simulations. To the best of the author's knowledge, 0.87 is the highest 30-dB rectangular factor among such wide fractional bandwidth THz BPF in the open literature. This high-performance filter is capable of improving the sideband rejection receiver performance.

Due to the electrical isolation nature of wireless power transfer (WPT) system, a method that can obtain the receiver side parameter only based on transmitter side information is desirable. However, typical estimation scheme assume that the coil inductance and compensation capacitance are constant, which will cause large estimation errors in high frequency system. Therefore, a multiple parameters estimation algorithm that can simultaneously identify load resistance, mutual inductance and receiver coil inductance is proposed in this paper, where only the input current and voltage are needed to be measured. Moreover, in order to cope with capacitor aging effect and manufacturing tolerance, compensation capacitance estimation algorithm is also presented. Mathematical model and algorithm procedure are depicted thoroughly and the gradient descent approach is adopted to obtain the optimal estimated parameter. The uniqueness of solution and the validity of this approach are strictly proved as well. Finally, a WPT experimental prototype with Class-D inverter is implemented and the experimental results are provided to verify the availability and accuracy of the proposed algorithm.

The V2C controlled buck converter by constant-frequency pulse-width modulation in continuous conduction mode gives rise to a great variety of instability behaviors, depending on the circuit parameter values. In this paper, the discrete-time model of the regulator is built by taking a current sampling resistor into consideration. The resulting Monodromy matrix is used to investigate the bifurcation phenomenon and stabilization property. The converter shows a series of period-doubling bifurcation phenomena accordingly as the feedback amplification coefficient increases. At the same time, a couple of Floquet multipliers pass through the unit circle along the negative real axis. Therefore, it reveals the mechanism of a series of period-doubling bifurcations happened in the system from the perspective of stability. Based on Floquet theory, a sine voltage compensation method is proposed to stabilize the bifurcation and chaotic behaviors. The simulations and experimental results proved the theoretical analysis.

An adaptive control law is proposed in this paper for body-fixed hovering over an asteroid with an unknown gravity field and an unknown rotation rate. The unknown parameters and irregular shape of an asteroid make the hovering problem challenging. To facilitate the controller design, first, it is assumed that the required parameters are attainable, and then a body-fixed hovering controller is designed using the backstepping approach. Based on the immersion and invariance theory, an adaptive law is devised to estimate the unknown parameters in conjunction with the backstepping controller. To overcome the integrability obstacle arising in the adaptive law design, the regressor matrix is decomposed to two matrices: a Jacobian matrix and a non-Jacobian matrix, and an auxiliary matrix is introduced into the latter to form a Jacobian matrix. Then, a bounded function is employed to dynamically scale the estimate errors and its dynamics is elaborately constructed to stabilize the system. A Lyaponuv function is constructed for the entire closed-loop system to achieve the stabilization conditions by LaSalle's theorem, and the bounded scaling function much simplifies the analysis. Finally, numerical simulations demonstrate the effectiveness of the proposed control scheme.

Directional modulation (DM) as a physical layer security technique has been studied based on the traditional antenna arrays; however, in most of the designs, only one signal is transmitted at one carrier frequency. In this paper, signal polarization information is exploited, and a new DM scheme is designed, which can transmit a pair of orthogonally polarized signals to the same direction at the same frequency simultaneously, resulting in doubled channel capacity. These two signals can also be considered as one composite signal using a four-dimensional (4-D) modulation scheme across the two polarization diversity channels. Moreover, compressive sensing (CS)-based formulations for designing sparse crossed-dipole arrays in this context are proposed to exploit the degrees of freedom in the spatial domain for further improved performance, as demonstrated by various design examples.

Traditional directional modulation designs are normally based on a single phased array, with the assumption that eavesdroppers and desired users are in different directions. However, it is not always the case, as it is possible that they are in the same direction. As a result, signals received by eavesdroppers will be approximately the same as desired users. To solve the problem, a multiple antenna array model is introduced, and the principle of the design is that the eavesdropper located in the same direction as the desired user for one antenna array may not be in the same direction for another antenna array. Based on such a model, in this paper, a power minimization constraint for the design is proposed, and an artificial noise component is included, resulting in a scrambled phase response for eavesdroppers. Beam pattern, phase pattern, and bit error rate are used as metrics to show the effectiveness of the designs based on multiple fixed antenna arrays and location optimized antenna arrays, respectively.

Urban air pollutant concentration prediction is dealing with a surge of massive environmental monitoring data and complex changes in air pollutants. This requires effective prediction methods to improve prediction accuracy and to prevent serious pollution incidents, thereby enhancing environmental management decision-making capacity. In this paper, a new pollutant concentration prediction method is proposed based on the vast amounts of environmental data and deep learning techniques. The proposed method integrates big data by using two kinds of deep networks. This method is based on the design that uses a convolutional neural network as the base layer, automatically extracting features of input data. A long short-term memory network is used for the output layer to consider the time dependence of pollutants. Our model consists of these two deep networks. With performance optimization, the model can predict future particulate matter (PM2.5) concentrations as a time series. Finally, the prediction results are compared with the results of numerical models. The applicability and advantages of the model are also analyzed. The experimental results show that it improves prediction performance compared with classic models.

This paper addressed the vessel segmentation and disease diagnostic in coronary angiography image and proposed an Encoder-Decoder architecture of deep learning with End-to-End model, where Encoder is based on ResNet, and the deep features are exacted automatically, and the Decoder produces the segmentation result by balanced cross-entropy cost function. Furthermore, batch normalization is employed to decrease the gradient vanishing in the training process, so as to reduce the difficulty of training the deep neural network. The experiment results show that the algorithm effectively exacts the feature and edge information, therefore the complex background disturbance is suppressed convincingly, and the vessel segmentation precision is improved effectively, the segmentation precision for three typical vessels are 0.8365, 0.8924 and 0.6297 respectively; and the F-measure are 0.8514, 0.8786 and 0.7298, respectively. In addition, the experiment results show that our proposed can be generalized to the angiography image within limits.

Time-lapse is a technology used to record the development of embryos during in-vitro fertilization (IVF). Accurate classification of embryo early development stages can provide embryologists valuable information for assessing the embryo quality, and hence is critical to the success of IVF. This paper proposes a multi-task deep learning with dynamic programming (MTDL-DP) approach for this purpose. It first uses MTDL to pre-classify each frame in the time-lapse video to an embryo development stage, and then DP to optimize the stage sequence so that the stage number is monotonically non-decreasing, which usually holds in practice. Different MTDL frameworks, e.g., one-to-many, many-to-one, and many-to-many, are investigated. It is shown that the one-to-many MTDL framework achieved the best compromise between performance and computational cost. To our knowledge, this is the first study that applies MTDL to embryo early development stage classification from time-lapse videos.

A novel compact quasi-planar four-way power divider with a wide isolation bandwidth is proposed in this paper. The signal is fed through the coplanar waveguide and then divided to the four microstrip-line output ports through a metal via. To realize the high output isolation, the multi-layer circuit structure is used. The isolation network is composed of four isolation resistors and a metal via (common node). Equivalent circuits are given to analyze and design the novel power divider. The detailed analyses are given according to the equivalent circuits. A four-way high-isolation power divider with the center frequency of 2.1 GHz is designed and fabricated. The measured 20-dB relative isolation bandwidth between the output ports is about 58%, while 30-dB relative isolation bandwidth between the output ports is about 10%. The measured results agree well with the simulated ones. The total size of the fabricated four-way power divider is 0.17λg × 0.22λg (λg is the wavelength of the center frequency). It shows that the proposed power divider has not only the advantages of high output isolation and compact size but also the advantages of excellent output and input impedance matching and very low insertion loss. The presented power divider can be used in antenna feed, power-combining amplifier, and so on.

Bearing faults are the most common failure modes in the rotating system. Vibration data from the rotating system carry important information, that is, characterization and diagnosis; therefore, the vast vibration signals collected from multiple sensors mounted in different sites are transmitted in a certain order for online fault diagnosis. However, due to the influence of transfer paths and noises, the sensitivities to the same fault signal of measured data streams are of significant differences, and signals containing weak sensitivity to the fault are likely to be transmitted preferentially while neglecting transmission order. Meanwhile, high volume vibration data greatly increase the transmission burden. These above-mentioned reasons dramatically reduce online diagnostic efficiency. Thus, fully considering the sensitive differences to the fault for multiple channels, how to transmit measured data streams of multiple sensors for timely online detecting the bearing failure is still a primary challenge. In order to solve this problem, a novel online bearing fault diagnosis method based on the multiple data streams transmission schemes (MDSTS) is proposed in this paper. Multiple sensors are numbered consecutively, and data streams from all channels are transmitted according to the preset order and transport protocol via a certain length at the beginning of diagnosis. Then, a fault sensitivity assessment model (FSAM) is established on maximum mean discrepancy (MMD) for transmitting the most sensitive data stream by calculating the distribution discrepancies between each channel's data streams and the historical datasets in the frequency domain, and then, the fault diagnosis model based on K-nearest neighbor (KNN) trained on historical datasets was used to evaluate the transmission scheme and acquire reliable diagnostic results via predicting performances of multiple and consecutive datablocks until all these exceed an alarm value. The extensive experiment results show that the proposed method can timely and accurately identify the bearing faults and outperforms obviously competitive approaches.

Hovering over an asteroid is challenging due to both the large gravitational uncertainty and the strong external disturbance. When a spacecraft is not equipped with velocity sensors to simplify the navigation system and reduce the mass and costs, or the velocity sensors malfunction, the situation may be worse. In this paper, an extended state observer (ESO) is devised to estimate the velocity and the lumped disturbance simultaneously. The estimate errors are proved to be uniformly ultimately bounded with the assumption that the change rate of the lumped disturbance is bounded. To cope with control input saturation, a backstepping controller in conjunction with an auxiliary system is proposed for the hovering. Combining the controller with the ESO, we develop a disturbance rejection control scheme without velocity feedback. The stability of the whole closed-loop system is analyzed via the Lyapunov theory, showing that the control scheme can drive a spacecraft to the neighborhood of the desired hovering state. The numerical simulations of both body-fixed hovering and inertial hovering are conducted to demonstrate the effectiveness of the proposed scheme.

Providing accurate real-time environmental monitoring services is a meaningful but difficult task. Whereas recent developments in the Internet of Things and the fog computing paradigm have brought new opportunities to improve the service, achieving accurate environmental monitoring faces several challenges. First, data collected by the Internet of Things varies in temporal-spatial distribution, quality, and relevance to objectives. Second, due to the limited communication cost and stability, local real-time monitoring systems can be hardly achieved by adopting centralized cloud paradigm solutions in the real world. Third, the amount of data at a single edge node of fog computing paradigm is usually not sufficient for accurate environmental monitoring. In this paper, we propose a framework for environmental monitoring based on fog computing that uses multi-source heterogeneous data collected from the Internet of Things sensors. At each edge node, we employ local sub-classifiers to analyze the collected data and afterwards, a deep neural network based model to aggregate results from sub-classifiers. For the homologous sub-classifiers at different edge nodes, we design a federated learning mechanism to update sub-classifiers in concert by model transmission. We use multi-source heterogeneous data collected in Beijing to evaluate the proposed fog computing framework. Experimental results show that our federated learning mechanism has almost the same performance compared with centralized data fusion mechanism when training rounds are increased.

This paper considers the bipartite consensus of linear multi-agent systems with measurement noise, in which both the cooperation and competition relations exist among agents. To attenuate the influence of measurement noises, an important function α(t) is introduced. Then, the output-based controller with both the noisy part and noise-free part is proposed. It is proved that the proposed controller can guarantee the mean-square bipartite consensus with the assumption in which the signed graph is connected and structurally balanced. Meanwhile, the obtained result releases the limitation on time-varying function α(t). Finally, an example with six agents is utilized to illustrate the effectiveness of the proposed controller.