Search

PubMed ID: 32392253, In this work, a novel sensor based on printed circuit board (PCB) microstrip rectangular patch antenna is proposed to detect different ratios of ethanol alcohol in wines and isopropyl alcohol in disinfectants. The proposed sensor was designed by finite integration technique (FIT) based high-frequency electromagnetic solver (CST) and was fabricated by Proto Mat E33 machine. To implement the numerical investigations, dielectric properties of the samples were first measured by a dielectric probe kit then uploaded into the simulation program. Results showed a linear shifting in the resonant frequency of the sensor when the dielectric constant of the samples were changed due to different concentrations of ethanol alcohol and isopropyl alcohol. A good agreement was observed between the calculated and measured results, emphasizing the usability of dielectric behavior as an input sensing agent. It was concluded that the proposed sensor is viable for multipurpose chemical sensing applications. © 2020 Karatepe et al.

In order to improve both the discriminative power for models' local parts and the searching efficiency in 3D CAD model retrieval, a novel hierarchical feature descriptor for retrieval based on spatial bag of words is proposed in this paper. By extracting the essential information from Boundary Representation (B-Rep), 3D CAD models are transformed to Labelled Attribute Adjacency Graphs (LAAGs). Next, the models in training dataset are segmented into different regions according to their corresponding LAAG with an improved segmentation method. All collections of these local regions are described as local feature vectors with graph spectrum, and the codebook is created by clustering all these vectors. Each library model is then decomposed with the same methods mentioned above and globally represented as a spatial histogram of word pairs along with the adjacent relations of its regions, called Spatial Bags-of-Words (SBoWs),and then, the hierarchical feature descriptor(HFD) of each library model composed of global SBoWs and local graph spectrum is constructed. Finally, according to HFD, a two-level searching framework is presented for CAD model retrieval: the candidates are acquired by comparing the query with each target model based on their SBoWs vectors, and the remaining candidates are verified using optimal matching algorithm according to their local features. Experimental results show that the proposed methods promote both retrieval quality and efficiency significantly, so they can support the effective reuse of CAD models for enterprises.

In order to discriminate the real targets, the clutter and the dense multi-false targets, we propose a factorized convolutional neural network-based algorithm for radar targets discrimination. We establish the factorized convolutional neural network model with depthwise separable convolution. To reduce the parameters of the model, we establish the simplified factorized convolutional neural network by reducing the numbers of both convolutional filters and connection nodes of fully connected layers. The result of the measured data demonstrates that, as compared with the existing model, the simplified factorized convolutional neural network has higher discrimination rate for the real targets, the clutter and the dense multi-false targets, and its parameters are less than ten percent counterpart of a recent proposed model.

Graphene modified dye sensitized solar cells (DSSCs) show high photovoltaic performances due to the fast electron transport capacity and the large BET area of the additive. However, the corresponding research on how to further improve the properties of DSSCs by optimizing the adopted graphene is insufficient. In this study, reduced graphene oxide (RGO) nanosheets and three-dimensional graphene networks (3DG) are employed to build high performance photoanodes with a two-layer structure by taking full advantages of them, simultaneously. Residual functional groups of RGO support preconditions to construct an electron transport layer between conducting substrate and work layer of photoanode, which is beneficial to depress dark current on the interface. On the other hand, high integrity and continuity of 3DG provide a fast electrons transport network in the work layer, which suppresses the recombination of electron–hole pairs in the photoanode. After optimizing the structure of photoanode, the energy conversion efficiency reaches 8.87%.

Graphene is an ideal two-dimensional crystal with the advantages of high conductivity, unique physical and chemical properties, and high specific surface area. Especially, because of its super excellent electronic properties, graphene may substitute the traditional semiconductor silicon material and carbon nanotube, thus creating a new nanoscale electronic device. In addition, multilayer graphene with ultra-wide spectral absorption characteristics and unique photoelectric properties is an ideal material for photovoltaic devices. However, the zero band gap and semi-metality of graphene both limit its application in space detectors such as the microelectronic industries and satellites. Opening and regulating the graphene band gap by physical methods has become one of the key means to further expand its applications. Research work has shown that the doping of elements can significantly change the electronic structure of graphene, thereby regulating the optical properties of graphene. In order to provide an insight into electronic properties of graphene and tune its electronic band structure and optical properties effectively, electronic and optical properties of Ni-doped multi-layer graphene are studied and a number of interesting results are obtained. The calculation are carried out by the CASTEP tool in Materials Studio software based on the first-principles of ultrasoft pseudopotential of density functional theory. The models of three typical doping positions relative to carbon atoms are constructed. After structural optimization, it is obtained that " above the center of two carbon atoms” is the most stable doping structure. By using the method of local density approximation, the band structure, density of states, dielectric constant, reflectivity and refractive index of the models are calculated. The results show that an enhanced energy band gap can be achieved after nickel-doping, and reach up to 0.604 eV. Besides, peaked phenomenon of density of states at Femi level can be observed, which is accomplished by enhancing the plasma energy. Furthermore, the calculations show that the imaginary part of permittivity and refractive index increase after nickel-doping, suggesting that the optical absorbing performance is improved. All these results provide theoretical guidance for further exploring the optical properties of graphene.

High-performance absorbing material can play an important role in electromagnetic compatibility, electromagnetic radiation protection, and anti-detection of special equipment. Combining traditional absorbing material with metamaterial is an important direction for developing absorbing material. The composite absorbing body based on the development of metamaterial has advantages of thin thickness, light weight, strong absorption, and adjustable absorption band, but the super material absorption body composed of single-sized metal pattern elements possesses generally strong absorption only for electromagnetic waves at a certain frequency. It is difficult to meet the requirement for wide frequency absorption in practical applications. In order to broaden the absorption bandwidth of metamatial, metal spiral-ring metasurface coated short carbon fiber absorber with enhanced microwave absorbing performance is proposed. The absorber is a two-dimensional structure formed by periodically arranging a large number of individual absorber units in the horizontal and vertical direction. In the HFSS simulation software, a " master-slave boundary condition” consisting of " master boundary” and " slave boundary” is provided. Under this boundary condition, the electric field between adjacent boundaries has a phase difference, which can be used to simulate an infinite array. The research results show that the obvious enhancement of both the absorption peak and bandwidth can be observed by embedding the double-layer spiral-ring metasurfaces. The increase of initial length of spiral-rings and thickness of absorber are beneficial to further enhancing the microwave absorption. The reflection loss from 9.2 GHz to 18.0 GHz are under –10 dB (the bandwidth reaches 8.8 GHz), and the peak of S11 is –14.4 dB. Besides, we find that the effective electromagnetic parameters and impedance of spiral-ring metasurface embedded microwave absorber present obvious resonant phenomenon at multi-frequencies by calculating S parameters. Furthermore, an equivalent circuit model regarding double-layer spiral-ring embedded absorber is established to reveal the attenuation mechanism of microwave energy. The resonant frequencies derived from this model are well accord with the simulated results. Thereby, the multi-electromagnetic resonant frequencies make the composite microwave absorber combined with double-layer metal spiral-ring and carbon fiber have microwave reflection loss in a wide bandwidth.

In recent years, amorphous InGaZnO thin-film transistor (InGaZnO TFT) has attracted intensive attention. Due to its high mobility, low off-state current, and excellent uniformity over large fabrication area, the InGaZnO TFTs promise to replace silicon-based TFTs in flat panel displays, optical image sensors, touch sensing and fingerprint sensing area. The on-state performances of InGaZnO TFT are used in thin film transistor liquid crystal display, active-matrix organic light emitting display, etc. Consequently, numerous on-current models have been proposed previously. However, for lots of the emerging sensing applications such as optical image sensors, the leakage current of InGaZnO TFTs is critical.Previous literature has shown that the leakage current generation mechanisms in TFTs include trap-assisted thermal emission, trap-assisted field emission, inter-band tunneling, and auxiliary thermal electron field emission containing Poole-Frenkel effect. However, up to now, there has been few reports on the leakage current model of InGaZnO TFT, which hinders further the development of emerging applications in InGaZnO TFTs for sensor and imagers integrated in display panels.In this paper, the leakage current model of InGaZnO TFT is established on the basis of carrier generation recombination rate. The feasibility of the proposed model is proved by comparing the TCAD simulations with the measured results. In addition, the influences of geometrical parameters on the leakage current of InGaZnO TFT, i.e. the channel width, the active layer thickness, and the gate dielectric thickness, are analyzed in detail. This research gives insightful results for designing the sensors and circuits by using the InGaZnO TFTs.

The purpose of reverse engineering is to convert a large points cloud into a CAD model. Parameters extraction for rotational surface from measure data is an important problem in reverse engineering. Rotational axis is a crucial parameter of rotational surface. In this paper, random sampling consensus algorithm combined with Least-Squares method is used to extract the rotational axis of rotational surface. Experimental results show effectiveness, robustness and accuracy of this proposed approach.

The purpose of reverse engineering is to convert a large points cloud into a CAD model. Parameters extraction for extruded surface from measure data is an important problerm in reverse engineering. Extruded direction is a crucial parameter of extruded surface. In this paper, random sampling consensus algorithm combined with Least-Squares method is used to extract the extruded direction of extruded surface. Experimental results show effectiveness, robustness and accuracy of this proposed approach.

In reinforcement learning (RL), the exploration/exploitation (E/E) dilemma is a very crucial issue, which can be described as searching between the exploration of the environment to find more profitable actions, and the exploitation of the best empirical actions for the current state. We focus on the single trajectory RL problem where an agent is interacting with a partially unknown MDP over single trajectories, and try to deal with the E/E in this setting. Given the reward function, we try to find a good E/E strategy to address the MDPs under some MDP distribution. This is achieved by selecting the best strategy in mean over a potential MDP distribution from a large set of candidate strategies, which is done by exploiting single trajectories drawn from plenty of MDPs. In this paper, we mainly make the following contributions: (1) We discuss the strategy-selector algorithm based on formula set and polynomial function. (2) We provide the theoretical and experimental regret analysis of the learned strategy under an given MDP distribution. (3) We compare these methods with the “state-of-the-art” Bayesian RL method experimentally.

In recent years, considerable attention has been paid to amorphous indium gallium-zinc-oxide (a-IGZO) thin film transistors (TFTs) for high performance flat panel display, such as liquid-crystal displays (LCDs), active-matrix organic light-emitting diode (AMOLED) display and flexible display. This is because IGZO TFTs are more suitable for pixels and circuit integrations on display panel than the conventional silicon-based devices. The merits of IGZO TFT technology include high mobility, decent reliability, low manufacturing cost, and excellent uniformity over large fabrication area. However, it was reported that the electrical characteristics of IGZO TFT are susceptible to shift after electrical aging measurement under illumination, which is caused by the activation of trapped electrons from sub-gap states to conducting states. Therefore, it is necessary to introduce light shielding layer to suppress the electrical characteristic shift under illumination aging measurements. Lim et al. demonstrated the characteristics of IGZO TFT with additional light shielding metal layer, and proved that the threshold voltage of TFT can be tuned linearly by adjusting the biasing voltage of the light shielding metal. Taking advantage of this tunable threshold voltage, AMOLED pixel circuit with a threshold voltage shift compensation function can be implemented. However, drawback of this method lies in the adding of additional biasing line, which increases the circuit area and restricts the integration of high-resolution pixel circuits. Thus, Zan et al. proposed adopting floating (unbiased) light shielding metal layer to improve the characteristics of device. However, Zeng et al. demonstrated the abnormal output characteristics of the IGZO TFT, as it cannot be saturated due to the introduction of floating light shielding metal layer. It seems that the IGZO TFT with floating metal is different from the conventional double-gate or single gate structure. To date, the current conducting mechanism of IGZO TFT with floating metal has not been discussed yet. In this paper, the distribution of electrical potential in the IGZO TFT with a cross sectional view is thoroughly analyzed. It is confirmed that the abnormal output characteristic of IGZO TFT is caused by the capacitive coupling between the floating gate and the drain electrode of the transistor. On the basis of the voltage distribution relationship between the equivalent capacitances, a threshold-voltage-dependent current-voltage model is proposed. The simulated results by technology computer-aided design tool and those by the proposed model are in good agreement with each other. Therefore, the mechanism of floating gate effect for IGZO TFT is comprehensively demonstrated. The illustrated conducting mechanism and the proposed current-voltage model are helpful in developing the device and process of IGZO TFT with novel structure.

Students learn in many ways. There are all kinds of Teaching methods. When a student's learning style is compatible with the instructor's teaching style, the assumed learning outcomes will be obtained. So a good intelligent E-Learning system needs adjust teaching style according to the students' learning style. In the paper, students' modeling in intelligent e-learning system is considered. The models are based on the learning styles dimensions of the Felder & Silverman model. Students are divided into groups by decomposition and clustering method. The process is under the control of parameters such as distance and maximal number of groups. Experimental results show that a effective students' modeling can be obtained by the method.

Real time Video transport over ad hoc network poses a great challenge due to the high variability of the network bandwidth. In this paper, to adjust rate in time according to the change of the quality of the channel of ad hoc network, we propose a scheme which judge the quality of the wireless channel proactively. The value of signal strength is used to be an important factor for judging the quality of wireless channel. Experiment results show that the averages of BER and PER can be reduced by employing this scheme.

TCP, originally designed for reliable, fixed networks, interprets all packet loss into network congestion. However, this never holds in ad hoc networks. Mobility, channel error or node exhaustibility induced errors may all lead to unnecessary energy costs in TCP. Our work is prompted by the lack of energy efficiency evaluation methods concerning the overall performance improvement, as opposed to partial enhancement in algorithm optimization. We exploit the energy consumption behaviors of SACK, Tahoe, Reno and NewReno option over ad hoc networks. Simulation results indicate that SACK out-performs other three options in the case of mobility and channel error while Tahoe, Reno and NewReno perform better than SACK when node exhaustibility happens. It is also observed that node exhaustibility affects the energy consuming behavior of the whole network most. Application of analytic hierarchy process (AHP) to those simulation results leads to the final recommendation of SACK as the most energy efficient option