[go: up one dir, main page]
Next Issue
Volume 8, June
Previous Issue
Volume 8, April
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
5.3
2.6
Journals
Electronics
Volume 8
Issue 5
share announcement

Electronics, Volume 8, Issue 5 (May 2019) – 125 articles

Cover Story (view full-size image): Identification of mathematical models is a critical step to efficiently design control structures for autonomous vehicles. In this paper, a number of regression techniques—ridge, kernel ridge, and symbolic regression—are used for the black-box identification of a surface marine vehicle. The aim is to develop general and robust mathematical models using real experimental data from random manoeuvres and trajectories. Results show that machine learning techniques are robust approaches to model surface marine vehicles, even providing interpretable results in closed form equations using techniques like symbolic regression. View this paper.
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
12 pages, 4007 KiB  
Article
Accelerometer-Based Gyroscope Drift Compensation Approach in a Dual-Axial Stabilization Platform
by Shutong Li, Yanbin Gao, Gong Meng, Gang Wang and Lianwu Guan
Electronics 2019, 8(5), 594; https://doi.org/10.3390/electronics8050594 - 27 May 2019
Cited by 12 | Viewed by 7618
Abstract
An accelerometer-based gyro drift compensation approach in a dual-axial stabilization platform is introduced in this paper. The stabilization platform consists of platform framework, drive motor, gyro and accelerometer module and contorl board. Gyro is an angular rate detecting element to achieve angular rate [...] Read more.
An accelerometer-based gyro drift compensation approach in a dual-axial stabilization platform is introduced in this paper. The stabilization platform consists of platform framework, drive motor, gyro and accelerometer module and contorl board. Gyro is an angular rate detecting element to achieve angular rate and rotation angle of the dynamic platform system. However, the platform system has an unstable factor because of the drift of gyro. The main contribution of this paper is to implement a convenient gyro drift compensation approach by using the accelerometer. In contrast to a kalman filtering method, this approach is simpler and practical due to the high-precision characteristic of the accelerometer. Data filtering algorithm and limit of threshold setting of total acceleration values are applied in this approach. The validity and feasibility of the proposed approach are evaluated by four tests under various conditions. Full article
(This article belongs to the Section Microelectronics)
►▼ Show Figures

Figure 1

Figure 1
<p>Coordinate system of two-axis stabilized platform.</p> Full article ">Figure 2
<p>Relation between carrier coordinate with heading coordinate.</p> Full article ">Figure 3
<p>Relation between pitch coordinate with heading coordinate.</p> Full article ">Figure 4
<p>Accelerometer compensation algorithm.</p> Full article ">Figure 5
<p>Tri-Axis inertial sensor.</p> Full article ">Figure 6
<p>Stabilized platform structure.</p> Full article ">Figure 7
<p>Experimental schemes.</p> Full article ">Figure 8
<p>Experiment 1 (Red line is angular output of gyro. Green line is angular output of accelerometer. Carmine line is angular output after compensation. Blue line is angular output after Kalman filter.). (<b>a</b>) Angular outputs of gyro/accelerometer; (<b>b</b>) Applying the proposed compensation method; (<b>c</b>) Applying Kalman filtering method; (<b>d</b>) Comparison of two methods.</p> Full article ">Figure 9
<p>Experiment 2 (Red line is angular output of gyro. Green line is angular output of accelerometer. Carmine line is angular output after compensation. Blue line is angular output after Kalman filter.). (<b>a</b>) Angular outputs of gyro/accelerometer; (<b>b</b>) Applying the proposed compensation method; (<b>c</b>) Applying Kalman filtering method; (<b>d</b>) Comparison of two methods.</p> Full article ">Figure 10
<p>Experiment 3 (Red line is angular output of gyro. Green line is angular output of accelerometer. Carmine line is angular output after compensation. Blue line is angular output after Kalman filter.). (<b>a</b>) Angular outputs of gyro/accelerometer; (<b>b</b>) Applying the proposed compensation method; (<b>c</b>) Applying Kalman filtering method; (<b>d</b>) Comparison of two methods.</p> Full article ">Figure 11
<p>Experiment 4 (Red line is angular output of gyro. Green line is angular output of accelerometer. Carmine line is angular output after compensation. Blue line is angular output after Kalman filter.). (<b>a</b>) Angular outputs of gyro/accelerometer; (<b>b</b>) Applying the proposed compensation method; (<b>c</b>) Applying Kalman filtering method; (<b>d</b>) Comparison of two methods.</p> Full article ">
12 pages, 963 KiB  
Article
A 30–40 GHz CMOS Receiver Front-End with 5.9 dB NF and 16.5 dB Conversion Gain for Broadband Spectrum Sensing Applications
by Hyunki Jung, Dzuhri Radityo Utomo, Saebyeok Shin, Seok-Kyun Han, Sang-Gug Lee and Junsung Kim
Electronics 2019, 8(5), 593; https://doi.org/10.3390/electronics8050593 - 27 May 2019
Cited by 5 | Viewed by 4268
Abstract
A broadband receiver front-end with low noise figure and flat conversion gain response is presented in this paper. The receiver front-end is a part of the broadband spectrum sensing receiver and processes 30–40 GHz of broad input spectrum followed by down-conversion to DC-10 [...] Read more.
A broadband receiver front-end with low noise figure and flat conversion gain response is presented in this paper. The receiver front-end is a part of the broadband spectrum sensing receiver and processes 30–40 GHz of broad input spectrum followed by down-conversion to DC-10 GHz of IF signal. The proposed work is comprised of a low noise amplifier (LNA), on-chip passive Balun, down conversion mixer, and output buffer. To achieve front-end target specification over 10 GHz input bandwidth, the stagger-tuned LNA is employed and the down conversion mixer is loaded with a 3rd-order LC ladder low pass filter. The prototype chip was implemented in 45 nm CMOS technology. The chip achieves 10.3–16.5 dB conversion gain, 5.9 dB integrated NF, and −11 dBm IIP3 from 30 to 40 GHz. The chip is realized within 0.42 mm 2 and consumes 96 mW from a 1.2 V supply. Full article
(This article belongs to the Section Microwave and Wireless Communications)
►▼ Show Figures

Figure 1

Figure 1
<p>Receiver front-end architecture: stagger-tuned two-stage low noise amplifier (LNA), down-conversion mixer loaded with 3-stage LC ladder filter, LO buffer, and output buffer.</p> Full article ">Figure 2
<p>Schema of the LNA.</p> Full article ">Figure 3
<p>Schema of the LNA’s first stage.</p> Full article ">Figure 4
<p>The condition of R<math display="inline"><semantics> <msub> <mrow/> <mi>D</mi> </msub> </semantics></math> for LNA input matching and stability.</p> Full article ">Figure 5
<p>The simulated frequency response of the LNA with stagger tuning.</p> Full article ">Figure 6
<p>Schema of the double-balanced mixer.</p> Full article ">Figure 7
<p>Schema of single-balanced mixer.</p> Full article ">Figure 8
<p>Modeling of the mixer load with additional inductor for bandwidth extension.</p> Full article ">Figure 9
<p>Simulated conversion gain of the down-conversion mixer with and without series peaking inductor L<math display="inline"><semantics> <msub> <mrow/> <mi>P</mi> </msub> </semantics></math>.</p> Full article ">Figure 10
<p>Schema of the two-stage LO buffer with reactive matching and a Smith chart with the trajectory from Z<math display="inline"><semantics> <msub> <mrow/> <mrow> <mi>B</mi> <mi>U</mi> <mi>F</mi> </mrow> </msub> </semantics></math> to the 50 <math display="inline"><semantics> <mo>Ω</mo> </semantics></math>.</p> Full article ">Figure 11
<p>Chip photograph of the proposed receiver front-end.</p> Full article ">Figure 12
<p>Measured (<b>a</b>) input reflection coefficient (S<math display="inline"><semantics> <msub> <mrow/> <mn>11</mn> </msub> </semantics></math>), (<b>b</b>) conversion gain and NF over a 10 GHz IF frequency range; (<b>c</b>) IIP3 with frf1 = 35 GHz and frf2 = 35.1 GHz; and (<b>d</b>) 1 dB compression point with frf = 35 GHz.</p> Full article ">
17 pages, 1018 KiB  
Article
EEkNN: k-Nearest Neighbor Classifier with an Evidential Editing Procedure for Training Samples
by Lianmeng Jiao, Xiaojiao Geng and Quan Pan
Electronics 2019, 8(5), 592; https://doi.org/10.3390/electronics8050592 - 27 May 2019
Cited by 3 | Viewed by 2996
Abstract
The k-nearest neighbor (kNN) rule is one of the most popular classification algorithms applied in many fields because it is very simple to understand and easy to design. However, one of the major problems encountered in using the kNN [...] Read more.
The k-nearest neighbor (kNN) rule is one of the most popular classification algorithms applied in many fields because it is very simple to understand and easy to design. However, one of the major problems encountered in using the kNN rule is that all of the training samples are considered equally important in the assignment of the class label to the query pattern. In this paper, an evidential editing version of the kNN rule is developed within the framework of belief function theory. The proposal is composed of two procedures. An evidential editing procedure is first proposed to reassign the original training samples with new labels represented by an evidential membership structure, which provides a general representation model regarding the class membership of the training samples. After editing, a classification procedure specifically designed for evidently edited training samples is developed in the belief function framework to handle the more general situation in which the edited training samples are assigned dependent evidential labels. Three synthetic datasets and six real datasets collected from various fields were used to evaluate the performance of the proposed method. The reported results show that the proposal achieves better performance than other considered kNN-based methods, especially for datasets with high imprecision ratios. Full article
(This article belongs to the Special Issue Fuzzy Systems and Data Mining)
►▼ Show Figures

Figure 1

Figure 1
<p>A simplified three-class classification example.</p> Full article ">Figure 2
<p>Illustration of dependence among edited training samples.</p> Full article ">Figure 3
<p>Classification results for different combination rules under different <math display="inline"><semantics> <msub> <mi>k</mi> <mrow> <mi>e</mi> <mi>d</mi> <mi>i</mi> <mi>t</mi> </mrow> </msub> </semantics></math> values with values of <span class="html-italic">k</span> ranging from 1–25.</p> Full article ">Figure 4
<p>Classification results of the EE<span class="html-italic">k</span>NN classifier for different values of <math display="inline"><semantics> <msub> <mi>k</mi> <mrow> <mi>e</mi> <mi>d</mi> <mi>i</mi> <mi>t</mi> </mrow> </msub> </semantics></math> and <span class="html-italic">k</span>.</p> Full article ">Figure 5
<p>Training set and classification results for Case 1 with imprecision ratio <math display="inline"><semantics> <mrow> <mi>ρ</mi> <mo>=</mo> <mn>33</mn> <mo>%</mo> </mrow> </semantics></math>.</p> Full article ">Figure 6
<p>Training set and classification results for Case 2 with imprecision ratio <math display="inline"><semantics> <mrow> <mi>ρ</mi> <mo>=</mo> <mn>60</mn> <mo>%</mo> </mrow> </semantics></math>.</p> Full article ">Figure 7
<p>Training set and classification results for Case 3 with imprecision ratio <math display="inline"><semantics> <mrow> <mi>ρ</mi> <mo>=</mo> <mn>79</mn> <mo>%</mo> </mrow> </semantics></math>.</p> Full article ">Figure 8
<p>Classification results of different methods for real datasets.</p> Full article ">
19 pages, 7619 KiB  
Article
Active Disturbance Rejection Control of Multi-Joint Industrial Robots Based on Dynamic Feedforward
by Xin Cheng, Xiao Tu, Yunfei Zhou and Rougang Zhou
Electronics 2019, 8(5), 591; https://doi.org/10.3390/electronics8050591 - 27 May 2019
Cited by 14 | Viewed by 3758
Abstract
In this paper, the dynamics-based high-performance robot motion control technology has been mainly studied, and the overall structure is controlled via dynamics forward, given the nonlinearity, strong coupling and time-variability of robots. Considering the unavailability of precise robot model parameters and the uncertain [...] Read more.
In this paper, the dynamics-based high-performance robot motion control technology has been mainly studied, and the overall structure is controlled via dynamics forward, given the nonlinearity, strong coupling and time-variability of robots. Considering the unavailability of precise robot model parameters and the uncertain disturbance in real operation, we put forward an active disturbance rejection control (ADRC) strategy based on dynamic feedforward, aiming to improve the control robustness and combining the simple structure, strong anti- disturbance ability, and no restriction from the control model of ADRC. Given the multi-joint coupling of robots, controlled decoupling is conducted by using dynamic characteristics. The ADRC cascade control structure and algorithm based on dynamic feedforward have been studied and the closed-loop stability of the system is investigated by analyzing the system dynamic linearization compensation and the anti-disturbance ability of the extended state observer. Experiments have shown the new strategy is more robust over uncertain disturbance than the conventional proportional-integral-derivative control strategy. Full article
(This article belongs to the Special Issue Motion Planning and Control for Robotics)
►▼ Show Figures

Figure 1

Figure 1
<p>Model-based control structure. (<b>a</b>) dynamic feedback compensation; (<b>b</b>) dynamic feedforward.</p> Full article ">Figure 2
<p>Transmission structure from robot motor end to joint end.</p> Full article ">Figure 3
<p>Robot ADRC control decoupling.</p> Full article ">Figure 4
<p>Structure of ADRC feedforward control 2-order system.</p> Full article ">Figure 5
<p>Structure of the cascade system.</p> Full article ">Figure 6
<p>Active disturbance rejection control (ADRC) cascade controller based on inertia feedforward.</p> Full article ">Figure 7
<p>Experimental robot. (<b>a</b>) Robot end-point end P trajectory; (<b>b</b>) coordinate system.</p> Full article ">Figure 8
<p>Robot uniaxial PI controller.</p> Full article ">Figure 9
<p>The trajectory follow-up results and error curve of joint 2 in different controllers (w = 2 rad/s). (<b>a</b>) trajectory follow-up curve of PI controller; (<b>b</b>) trajectory following errors of PI controller; (<b>c</b>) trajectory follow-up curve of ADRC controller; (<b>d</b>) trajectory following error of ADRC controller based on dynamic feedforward.</p> Full article ">Figure 10
<p>Robot end-point Cartesian follow-up trajectory using ADRC controller (w = 2 rad/s). (<b>a</b>) 3D image; (<b>b</b>) XOY planar projection; (<b>c</b>) XOZ planar projection; (<b>d</b>)YOZ planar projection.</p> Full article ">Figure 10 Cont.
<p>Robot end-point Cartesian follow-up trajectory using ADRC controller (w = 2 rad/s). (<b>a</b>) 3D image; (<b>b</b>) XOY planar projection; (<b>c</b>) XOZ planar projection; (<b>d</b>)YOZ planar projection.</p> Full article ">Figure 11
<p>Robot end-point Cartesian following trajectory using PI controller (w = 2 rad/s). (<b>a</b>) 3D image; (<b>b</b>) XOY planar projection; (<b>c</b>) XOZ planar projection; (<b>d</b>) YOZ planar projection.</p> Full article ">Figure 12
<p>Robot end-point Cartesian feedforward forward controller at different speeds. (<b>a</b>) 3D image; (<b>b</b>) XOY planar projection; (<b>c</b>) XOZ planar projection; (<b>d</b>) YOZ planar projection.</p> Full article ">Figure 12 Cont.
<p>Robot end-point Cartesian feedforward forward controller at different speeds. (<b>a</b>) 3D image; (<b>b</b>) XOY planar projection; (<b>c</b>) XOZ planar projection; (<b>d</b>) YOZ planar projection.</p> Full article ">
12 pages, 851 KiB  
Article
Secure and Efficient Data Sharing Scheme Based on Certificateless Hybrid Signcryption for Cloud Storage
by Wei Luo and Wenping Ma
Electronics 2019, 8(5), 590; https://doi.org/10.3390/electronics8050590 - 27 May 2019
Cited by 24 | Viewed by 3696
Abstract
As cloud service providers are not completely trusted, people are increasingly concerned about security issues such as data confidentiality and user privacy. In many existing schemes, the private key generator (PKG) generates a full private key for each user, which means that the [...] Read more.
As cloud service providers are not completely trusted, people are increasingly concerned about security issues such as data confidentiality and user privacy. In many existing schemes, the private key generator (PKG) generates a full private key for each user, which means that the PKG can forge a valid signature or decrypt the ciphertext. To address the issue, we first present a novel certificateless hybrid signcryption (CL-HSC) scheme without pairing, in which the PKG only generates the partial private keys for users. It is provably secure under the Elliptic Curve Computational Diffie-Hellman (EC-CDH) assumption in the random oracle model. Then, we propose a key derivation method by which the data owner only needs to maintain the master key to get rid of the complex key management. By combining our proposed CL-HSC scheme and the key derivation method, we present a secure and efficient data-sharing scheme for cloud storage, which can resist collusion attacks, spoofing attacks, and replay attacks and makes user revocation easier. In addition, compared with some existing schemes, our scheme has a lower computational complexity. Full article
(This article belongs to the Section Computer Science & Engineering)
►▼ Show Figures

Figure 1

Figure 1
<p>The system model.</p> Full article ">Figure 2
<p>The Data Encryption Keys Generation Algorithm.</p> Full article ">Figure 3
<p>The experiment results.</p> Full article ">
10 pages, 4312 KiB  
Article
A High-Speed Low-Power Divide-by-3/4 Prescaler using E-TSPC Logic DFFs
by Tianchen Shen, Jiabing Liu, Chunyi Song and Zhiwei Xu
Electronics 2019, 8(5), 589; https://doi.org/10.3390/electronics8050589 - 27 May 2019
Cited by 2 | Viewed by 5480
Abstract
A high-speed, low-power divide-by-3/4 prescaler based on an extended true single-phase clock D-flip flop (E-TSPC DFF) is presented. We added two more transistors and a mode control signal to the conventional E-TSPC based divide-by-4 divider to achieve the function of the divide-by-3/4 dual [...] Read more.
A high-speed, low-power divide-by-3/4 prescaler based on an extended true single-phase clock D-flip flop (E-TSPC DFF) is presented. We added two more transistors and a mode control signal to the conventional E-TSPC based divide-by-4 divider to achieve the function of the divide-by-3/4 dual modulus frequency divider. The designed divide-by-3/4 achieved higher speed and lower power operation with mode control compared with the conventional ones. The prescaler was comprised of sixteen transistors and integrates an inverter in the second DFF to provide output directly. The power consumption was minimized due to the reduced number of stages and transistors. In addition, the prescaler operating speed was also improved due to a reduced critical path. We compared the simulation results with conventional E-TSPC based divide-by-3/4 dividers in the same process, where the figure-of-merit (FoM) of the proposed divider was 17.4–75.5% better than conventional ones. We have also fabricated the prescaler in a 40 nm complementary metal oxide semiconductor (CMOS) process. The measured highest operating frequency was 9 GHz with 0.303 mW power consumption under 1.35 V power supply, which agrees with the simulation well. The measurement results demonstrate that the proposed divider achieves high-speed and low-power operation. Full article
(This article belongs to the Special Issue Energy Efficient Circuit Design Techniques for Low Power Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Block diagram of the proposed phase-locked-loop (PLL) synthesizer.</p> Full article ">Figure 2
<p>Conventional divide-by-3/4 divider prescalers (<b>a</b>) not-or(NOR)-based prescaler; (<b>b</b>) multiplex(MUX)-based prescaler; (<b>c</b>) prescaler presented in Reference [<a href="#B12-electronics-08-00589" class="html-bibr">12</a>].</p> Full article ">Figure 3
<p>Proposed divide-by-3/4 frequency divider.</p> Full article ">Figure 4
<p>Waveforms of the proposed prescaler in divide-by-3 mode.</p> Full article ">Figure 5
<p>Simulation results of the proposed prescaler operating at 7.2 GHz @ 1.1 V supply and room temperature.</p> Full article ">Figure 6
<p>Die photo of the divider test chip.</p> Full article ">Figure 7
<p>The test setup. (<b>a</b>) The block diagram of the test setup; (<b>b</b>) photo of the test setup.</p> Full article ">Figure 8
<p>The measured waveforms of (<b>a</b>) divide-by-3 at 6.9 GHz under 1.1 V; (<b>b</b>) divide-by-4 at 6.9 GHz under 1.1 V; (<b>c</b>) divide-by-3 at 9 GHz under 1.35V; (<b>d</b>) divide-by-4 at 9 GHz under 1.35 V.</p> Full article ">
20 pages, 3477 KiB  
Article
Jet Features: Hardware-Friendly, Learned Convolutional Kernels for High-Speed Image Classification
by Taylor Simons and Dah-Jye Lee
Electronics 2019, 8(5), 588; https://doi.org/10.3390/electronics8050588 - 27 May 2019
Cited by 3 | Viewed by 3682
Abstract
This paper explores a set of learned convolutional kernels which we call Jet Features. Jet Features are efficient to compute in software, easy to implement in hardware and perform well on visual inspection tasks. Because Jet Features can be learned, they can be [...] Read more.
This paper explores a set of learned convolutional kernels which we call Jet Features. Jet Features are efficient to compute in software, easy to implement in hardware and perform well on visual inspection tasks. Because Jet Features can be learned, they can be used in machine learning algorithms. Using Jet Features, we make significant improvements on our previous work, the Evolution Constructed Features (ECO Features) algorithm. Not only do we gain a 3.7× speedup in software without loosing any accuracy on the CIFAR-10 and MNIST datasets, but Jet Features also allow us to implement the algorithm in an FPGA using only a fraction of its resources. We hope to apply the benefits of Jet Features to Convolutional Neural Networks in the future. Full article
(This article belongs to the Special Issue Convolutional Neural Network Design and Hardware Implementation for Real-Time Vision Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>An example a separable filter. A 3 × 3 Gaussian kernel can be separated into two convolutions with smaller kernels.</p> Full article ">Figure 2
<p>The four basic kernels of all Jet Features. The top two kernels can be thought of as scaling or blurring factors. The bottom two perform derivatives in the either the x or y direction. Every Jet Feature is simply a series of convolutions with any number of each of these kernels. The order does not matter.</p> Full article ">Figure 3
<p>These examples demonstrate how the Gaussian kernel and Sobel kernels are examples of Jet Features. These 3 × 3 kernels can be broken down into a series of four convolutions with the two cell Jet Feature kernels. The Sobel kernels are similar to the Gaussian, but one of the scaling factors is replaced with a partial derivative.</p> Full article ">Figure 4
<p>An example of a hypothetical ECO Feature made up of three transforms. The top boxes represent the type of each transform. The boxes below show each transform’s associated parameters. The number of transforms, transform types and parameters of each transforms are randomly initialized and then evolved through a genetic algorithm.</p> Full article ">Figure 5
<p>An example of mutation (<b>top</b>) and crossover (<b>bottom</b>). Mutation will only change the parameters of a given ECO Feature. Crossover takes the first part of one feature and appends the latter part of another feature to it.</p> Full article ">Figure 6
<p>A example paring of an ECO Feature with a random forest classifier. Every ECO Feature is paired with its own classifier. Originally, perceptrons were used, but in our work, random forests are used which offer multiclass classification.</p> Full article ">Figure 7
<p>A system diagram of the original ECO Features Algorithm. Each classifier has its own ECO Feature transform. The outputs of each classifier are collected into a weighted summation to determine the final prediction.</p> Full article ">Figure 8
<p>How mutation (top) and crossover (bottom) are defined for Jet Features.</p> Full article ">Figure 9
<p>System diagram for the ECO Jet architecture. The Jet Features Unit computes every feature for a given multiscale local jet on an input image. A router connects only the ones that were selected during training to a corresponding random forest. The forests each vote on a final prediction.</p> Full article ">Figure 10
<p>Convolution units. The top unit shows only a single buffer needed for convolution along rows in the <span class="html-italic">x</span> direction. The bottom unit shows a large buffer used for convolution along the columns in the <span class="html-italic">y</span> direction.</p> Full article ">Figure 11
<p>Comparison of multiscale local jets. The maximum allowable variance in Gaussian blur and order of differentiation was limited by 15, 10 and 5. A fourth case was tested where variance in Gaussian blur was bounded by 5 and order of differentiation and was bounded by 1.</p> Full article ">Figure 12
<p>The array of convolutions in the ECO Jet Features unit. The first dimension consists of five scaling factor units (<math display="inline"><semantics> <mrow> <mi>σ</mi> <mo>≤</mo> <mn>5</mn> </mrow> </semantics></math>) that use addition convolutions as shown in <a href="#electronics-08-00588-f010" class="html-fig">Figure 10</a>. The second dimension consists of a single partial derivative factor for each of the outputs of the previous dimension and the input (<math display="inline"><semantics> <mrow> <mi>δ</mi> <mo>≤</mo> <mn>1</mn> </mrow> </semantics></math>). The angled blocks represent further convolutions, suggesting the third and forth dimensions.</p> Full article ">Figure 13
<p>The arrangement of forests and trees for ECO Jet Feature predictions. Each ECO Jet Feature drives in input of a random forest. Each random forest is made up of multiple decision trees. Predictions of tree are tabulated to make the decision of a single forest. The votes from forests are weighted and tabulated to form the prediction for the entire system.</p> Full article ">Figure 14
<p>The hardware structure of a decision tree. The node data unit holds the pixel value and pixel location information for the pixel each nodes is compared to. It outputs the data for the next pixel that will be streamed in. The node-pixel comparison unit waits for a match from the incoming pixel location from the input and the next pixel location from the node data unit. The pixel and node values are compared. The result is sent to the node structure unit. Once a leaf is activated in the tree structure, the prediction unit is signaled and a prediction is issued from the tree.</p> Full article ">Figure 15
<p>Comparison of test accuracy to the number of total nodes the random forests. Various number of trees, forests and depths of trees were tested.</p> Full article ">Figure 16
<p>Examples from the MNIST dataset (<b>top</b>) with handwritten digits. Examples from the CIFAR-10 dataset (<b>bottom</b>). Classes include airplane, bird, car, cat, deer, dog, frog, horse, ship and truck.</p> Full article ">Figure 17
<p>Examples from the BYU Fish dataset. Each image is of a different fish species.</p> Full article ">Figure 18
<p>Accuracy comparison between the original ECO Features algorithm and the ECO Jet Features algorithm on CIFAR-10. Once the models seem to converge, there is no evidence of lost accuracy in the ECO Jet Features model.</p> Full article ">Figure 19
<p>Accuracy comparison between the original ECO Features algorithm and the ECO Jet Features algorithm on MNIST. Once the models seem to converge, ECO Jet Features seem to have a slight edge in accuracy, about 0.3%.</p> Full article ">Figure 20
<p>Accuracy comparison between the original ECO Features algorithm and the ECO Jet Features algorithm on the BYU Fish dataset.</p> Full article ">
13 pages, 979 KiB  
Article
Tree Sampling for Detection of Information Source in Densely Connected Networks
by Taewon Min and Changhee Joo
Electronics 2019, 8(5), 587; https://doi.org/10.3390/electronics8050587 - 27 May 2019
Cited by 1 | Viewed by 2652
Abstract
We investigate the problem of source detection in information spreading throughout a densely-connected network. Previous works have been developed mostly for tree networks or applied the tree-network results to non-tree networks assuming that the infection occurs in the breadth first manner. However, these [...] Read more.
We investigate the problem of source detection in information spreading throughout a densely-connected network. Previous works have been developed mostly for tree networks or applied the tree-network results to non-tree networks assuming that the infection occurs in the breadth first manner. However, these approaches result in low detection performance in densely-connected networks, since there is a substantial number of nodes that are infected through the non-shortest path. In this work, we take a two-step approach to the source detection problem in densely-connected networks. By introducing the concept of detour nodes, we first sample trees that the infection process likely follows and effectively compare the probability of the sampled trees. Our solution has low complexity of O ( n 2 log n ) , where n denotes the number of infected nodes, and thus can be applied to large-scale networks. Through extensive simulations including practical networks of the Internet autonomous system and power grid, we evaluate our solution in comparison with two well-known previous schemes and show that it achieves the best performance in densely-connected networks. Full article
(This article belongs to the Special Issue Data-Driven Network Security and Privacy)
►▼ Show Figures

Figure 1

Figure 1
<p>An iteration of Algorithm 1. After edge <math display="inline"><semantics> <mrow> <mo stretchy="false">(</mo> <mi>u</mi> <mo>,</mo> <mi>w</mi> <mo stretchy="false">)</mo> </mrow> </semantics></math> is included in tree <span class="html-italic">T</span>, the D-GENfunction randomly picks <math display="inline"><semantics> <mrow> <mo stretchy="false">(</mo> <mi>x</mi> <mo>,</mo> <mi>u</mi> <mo stretchy="false">)</mo> </mrow> </semantics></math> from <math display="inline"><semantics> <mrow> <mo>{</mo> <mrow> <mo stretchy="false">(</mo> <mi>x</mi> <mo>,</mo> <mi>u</mi> <mo stretchy="false">)</mo> </mrow> <mo>,</mo> <mrow> <mo stretchy="false">(</mo> <msup> <mi>x</mi> <mo>′</mo> </msup> <mo>,</mo> <mi>u</mi> <mo stretchy="false">)</mo> </mrow> <mo>}</mo> </mrow> </semantics></math> and adds it in tree <span class="html-italic">T</span> with probability <math display="inline"><semantics> <msubsup> <mi>P</mi> <mi>x</mi> <mi>d</mi> </msubsup> </semantics></math>.</p> Full article ">Figure 2
<p>Tree generation under Sample-Tree Generation (STG). (<b>a</b>) An example network graph. Gray circles denote the set of infected nodes <math display="inline"><semantics> <msub> <mi>V</mi> <mi>I</mi> </msub> </semantics></math>. (<b>b</b>) A tree generated by STG. Given <span class="html-italic">v</span> as a root, a tree marked by thick edges is generated under STG. Each number on the edges denotes the order that the edge is added. Due to the detour at <math display="inline"><semantics> <mrow> <mo stretchy="false">(</mo> <mi>c</mi> <mo>,</mo> <mi>g</mi> <mo stretchy="false">)</mo> </mrow> </semantics></math>, it is a non-Breadth First Search (BFS) tree.</p> Full article ">Figure 3
<p>Performance of Sample-Tree-based Estimator (STE), Closeness Centrality (CC), and Infection Eccentricity (IE) in regular trees.</p> Full article ">Figure 4
<p>Performance of STE, CC, and IE in general trees.</p> Full article ">Figure 5
<p>Performance of STE, CC, and IE in <span class="html-italic">p</span>-edge networks.</p> Full article ">Figure 6
<p>Performance of STE, CC, and IE in <span class="html-italic">r</span>-disk networks.</p> Full article ">Figure 7
<p>Detection performance in the Internet Autonomous System (IAS) network.</p> Full article ">Figure 8
<p>Detection performance in the power grid network.</p> Full article ">
16 pages, 8545 KiB  
Article
A Low-Cost, High-Precision Method for Ripple Voltage Measurement Using a DAC and Comparators
by Jincheng Liu, Jiguang Yue, Li Wang, Chenhao Wu and Feng Lyu
Electronics 2019, 8(5), 586; https://doi.org/10.3390/electronics8050586 - 27 May 2019
Cited by 2 | Viewed by 4461
Abstract
As the core of electronic system, the switched-mode power supply (SMPS) will lead to serious accidents and catastrophes if it suddenly fails. According to the related research, the monitoring of ripple can acquire the health degree of SMPS indirectly. To realize low-cost, high-precision, [...] Read more.
As the core of electronic system, the switched-mode power supply (SMPS) will lead to serious accidents and catastrophes if it suddenly fails. According to the related research, the monitoring of ripple can acquire the health degree of SMPS indirectly. To realize low-cost, high-precision, and automatic ripple measurement, this paper proposes a new ripple voltage (peak-to-peak value) measuring scheme, utilizing a DAC and two high-speed comparators. Within this scheme, the DC component of SMPS output is blocked by a high-pass filter (HPF). Then, the filtered signal and the reference voltage from a DAC together compose the input of a high-speed comparator. Finally, output pulses of the comparator are captured by a microcontroller unit (MCU), which readjusts the output of the DAC by calculation, and this process is repeated until the DAC output is exactly equal to the peak (or valley) value of ripple. Moreover, in order to accelerate the measurement process, a peak estimation method is specially designed to calculate the output ripple peak (or valley) value of buck topology through merely two measurements. Then the binary search method is utilized to obtain a more exact value on the basis of estimative results. Additionally, an analysis of the measurement error of this ripple measurement system is executed, which shows that the theoretical error is less than 0.5% where the ripple value is larger than 500 mV. Furthermore, appropriate components are selected, and a prototype is manufactured to verify the validity of the proposed theory. Full article
(This article belongs to the Special Issue Signal Processing and Analysis of Electrical Circuit)
►▼ Show Figures

Figure 1

Figure 1
<p>The occurrence of ripple in switched-mode power supply (SMPS) (taking buck topology as an example).</p> Full article ">Figure 2
<p>Block diagram of peak measurement and some inner signals.</p> Full article ">Figure 3
<p>Waveform of key nodes in <a href="#electronics-08-00586-f002" class="html-fig">Figure 2</a>. (<b>a</b>) The DAC output, with constant voltage in each feedback cycle. (<b>b</b>) The HPF output, only maintaining an AC ripple. (<b>c</b>) The SMPS output, containing a high DC bias and an AC ripple. (<b>d</b>) The output of comparator, providing a periodic pulse in most cases.</p> Full article ">Figure 4
<p>HPF diagram, (<b>a</b>) the actual components, (<b>b</b>) the parasitic parameters of components.</p> Full article ">Figure 5
<p>Flowchart of the proposed algorithm.</p> Full article ">Figure 6
<p>The differences between binary search and the proposed method. (<b>a</b>) Binary search method, the requiring measurement cycles are equal to the DAC bits. (<b>b</b>) Proposed method, reducing the measurement cycles by peak estimation (step 1) and bound determination (step 2). (<b>c</b>) Further reduction of measurement time by cutting the waiting time in first several cycles.</p> Full article ">Figure 7
<p>The prototype.</p> Full article ">Figure 8
<p>The theoretical measurement errors of different methods.</p> Full article ">Figure 9
<p>The experimental environment.</p> Full article ">Figure 10
<p>The measurement differences for different amplitudes.</p> Full article ">Figure 11
<p>The measurement differences for different duty ratios.</p> Full article ">Figure 12
<p>The measurement differences for different frequencies.</p> Full article ">
16 pages, 3796 KiB  
Article
A Hybrid QoS-QoE Estimation System for IPTV Service
by Jaroslav Frnda, Jan Nedoma, Jan Vanus and Radek Martinek
Electronics 2019, 8(5), 585; https://doi.org/10.3390/electronics8050585 - 27 May 2019
Cited by 17 | Viewed by 4268
Abstract
The internet protocol television service (IPTV) has become a key product for internet service providers (ISP), offering several benefits to both ISP and end-users. Because packet networks based on internet protocol have not been prepared for time-sensitive services, such as voice or video, [...] Read more.
The internet protocol television service (IPTV) has become a key product for internet service providers (ISP), offering several benefits to both ISP and end-users. Because packet networks based on internet protocol have not been prepared for time-sensitive services, such as voice or video, packet networks have had to adopt several mechanisms to secure minimal transmission standards in the form of data stream prioritization. There are two commonly used approaches for video quality assessment. The first approach needs an original source for comparison (full-reference objective metrics), and the second one requires observers for subjective evaluation of video quality. Both approaches are impractical in real-time transmission because it is difficult to transform an objective score into a subjective quality perception, and on the other hand, subjective tests are not able to be performed immediately. Since many countries worldwide put IPTV on the same level as other broadcasting systems (e.g., terrestrial, cable, or satellite), IPTV services are subject to regulation by the national regulation authority. This results in the need to prepare service qualitative criteria and monitoring tools capable of measuring end-user satisfaction levels. Our proposed model combines the principles of both assessment approaches, which results in an effective monitoring solution. Therefore, the main contribution of the created system is to offer a monitoring tool able to analyze the features extracted from the video sequence and transmission system and promptly translate their impact into a subjective point of view. Full article
(This article belongs to the Section Computer Science & Engineering)
►▼ Show Figures

Figure 1

Figure 1
<p>The block diagram of the SSIM index metric.</p> Full article ">Figure 2
<p>Stimulus presentation in the absolute category rating (ACR) method.</p> Full article ">Figure 3
<p>Spatial information (SI) and temporal information (TI) values for UHD video sequences [<a href="#B19-electronics-08-00585" class="html-bibr">19</a>].</p> Full article ">Figure 4
<p>The whole procedure of creation and evaluation of testing video sequences.</p> Full article ">Figure 5
<p>Subjective and objective metric results for the Campfire (<b>a</b>) and Construction (<b>b</b>) scenes.</p> Full article ">Figure 5 Cont.
<p>Subjective and objective metric results for the Campfire (<b>a</b>) and Construction (<b>b</b>) scenes.</p> Full article ">Figure 6
<p>Subjective and objective metric results for the Runner (<b>a</b>) and Wood (<b>b</b>) scenes.</p> Full article ">Figure 6 Cont.
<p>Subjective and objective metric results for the Runner (<b>a</b>) and Wood (<b>b</b>) scenes.</p> Full article ">Figure 7
<p>The relative error distribution.</p> Full article ">Figure 8
<p>Four testing video sequences [<a href="#B19-electronics-08-00585" class="html-bibr">19</a>].</p> Full article ">Figure 9
<p>Correlation diagrams for H.264 (<b>a</b>) and H.265 (<b>b</b>).</p> Full article ">
12 pages, 1098 KiB  
Article
Zi-CAM: A Power and Resource Efficient Binary Content-Addressable Memory on FPGAs
by Muhammad Irfan, Zahid Ullah and Ray C. C. Cheung
Electronics 2019, 8(5), 584; https://doi.org/10.3390/electronics8050584 - 27 May 2019
Cited by 14 | Viewed by 4254
Abstract
Content-addressable memory (CAM) is a type of associative memory, which returns the address of a given search input in one clock cycle. Many designs are available to emulate the CAM functionality inside the re-configurable hardware, field-programmable gate arrays (FPGAs), using static random-access memory [...] Read more.
Content-addressable memory (CAM) is a type of associative memory, which returns the address of a given search input in one clock cycle. Many designs are available to emulate the CAM functionality inside the re-configurable hardware, field-programmable gate arrays (FPGAs), using static random-access memory (SRAM) and flip-flops. FPGA-based CAMs are becoming popular due to the rapid growth in software defined networks (SDNs), which uses CAM for packet classification. Emulated designs of CAM consume much dynamic power owing to a high amount of switching activity and computation involved in finding the address of the search key. In this paper, we present a power and resource efficient binary CAM architecture, Zi-CAM, which consumes less power and uses fewer resources than the available architectures of SRAM-based CAM on FPGAs. Zi-CAM consists of two main blocks. RAM block (RB) is activated when there is a sequence of repeating zeros in the input search word; otherwise, lookup tables (LUT) block (LB) is activated. Zi-CAM is implemented on Xilinx Virtex-6 FPGA for the size 64 × 36 which improved power consumption and hardware cost by 30 and 32%, respectively, compared to the available FPGA-based CAMs. Full article
(This article belongs to the Special Issue Emerging Applications of Recent FPGA Architectures)
►▼ Show Figures

Figure 1

Figure 1
<p>A 4 × 4 BCAM. (SL0, SL1, SL2, and SL3 are the four search lines; ML0, ML1, ML2, and ML4 are the four matchlines; Enc.: A priority encoder).</p> Full article ">Figure 2
<p>Dividing of the CAM design in two blocks to save power consumption. (Sw: Search word, Add: Output address where the search word is found).</p> Full article ">Figure 3
<p>Zi-CAM Architecture. (LB: lookup tables block, RB: RAM block, BSel: Block Selector, Sw: Input search word, Add: Output address).</p> Full article ">Figure 4
<p>Two major blocks of Zi-CAM architecture. (<b>a</b>) Lookup table (LUT) block (LB) for 8 ×10 Zi-CAM having eight pages with four LUTs each. (<b>b</b>) RAM block (RB) for 8 ×10 Zi-CAM.</p> Full article ">
17 pages, 575 KiB  
Article
Analysis and Design of Functional Device for Vehicular Cloud Computing
by Guilu Wu, Sha Li, Shujun Wang, Yutong Jiang and Zhengquan Li
Electronics 2019, 8(5), 583; https://doi.org/10.3390/electronics8050583 - 26 May 2019
Cited by 1 | Viewed by 3478
Abstract
Relay technology application becomes prevalent nowadays, as it can effectively extend the communication distance, especially for vehicular networks with a limited communication range. Combined with vehicular cloud (VC), transmission efficiency can be improved by offloading partial data. Hence, designing a vehicle relay algorithm [...] Read more.
Relay technology application becomes prevalent nowadays, as it can effectively extend the communication distance, especially for vehicular networks with a limited communication range. Combined with vehicular cloud (VC), transmission efficiency can be improved by offloading partial data. Hence, designing a vehicle relay algorithm and implementation embedded vehicle device is critical. In this paper, VC is considered to deal with the complexity computation in our proposed system model. Without a loss of generality, an end-to-end vehicle communication with one assisted vehicle is analyzed firstly on a transmission link based on VC. Here, the signal-to-noise ratio (SNR) on the receiving end and link outage probability is obtained to enhance the link reliability. The VC computing helps us further simplify computational complexity. Subsequently, an embedded vehicle-enabled device is designed to achieve the optimal path relay selection in realistic vehicular environments. In the functional device framework, we display an optimal path relay selection algorithm according to the link quality. Finally, the performance of the transmission link on the outage probability related with SNR is verified in the simulation results. Meanwhile, the effect of the relay gain is also analyzed. The application of a vehicle-enabled embedded device could improve the performance of vehicular networks. Full article
(This article belongs to the Special Issue Vehicular Networks and Communications)
►▼ Show Figures

Figure 1

Figure 1
<p>The system model where vehicle <span class="html-italic">S</span> attempts to transmit a message to vehicle <span class="html-italic">D</span>.</p> Full article ">Figure 2
<p>Vehicle device overall function framework.</p> Full article ">Figure 3
<p>A comparison between the analytical results for the Probability Density Function (PDF) of a full link and the simulation results in different signal-to-noise ratio (SNR) cases.</p> Full article ">Figure 4
<p>A Monte Carlo simulation for the different types of Amplify-and-Forward (AF) relay gain <span class="html-italic">A</span>.</p> Full article ">Figure 5
<p>Outage probability vs. different outage threshold <math display="inline"><semantics> <msub> <mi>γ</mi> <mrow> <mi>t</mi> <mi>h</mi> </mrow> </msub> </semantics></math> for <math display="inline"><semantics> <mi>γ</mi> </semantics></math> = 8 and 16, <math display="inline"><semantics> <mrow> <msup> <mi>A</mi> <mn>2</mn> </msup> <mo>=</mo> <mn>1</mn> <mo>/</mo> <msubsup> <mi>h</mi> <mn>1</mn> <mn>2</mn> </msubsup> </mrow> </semantics></math>, and <math display="inline"><semantics> <mrow> <msup> <mi>A</mi> <mn>2</mn> </msup> <mo>=</mo> <mn>1</mn> <mo>/</mo> <mrow> <mo>(</mo> <msubsup> <mi>h</mi> <mn>1</mn> <mn>2</mn> </msubsup> <mo>+</mo> <msub> <mi>N</mi> <mn>0</mn> </msub> <mo>)</mo> </mrow> </mrow> </semantics></math>.</p> Full article ">Figure 6
<p>A comparison of the outage probability of AF relay-assisted vehicular networks based on Equation (<a href="#FD22-electronics-08-00583" class="html-disp-formula">22</a>).</p> Full article ">Figure 7
<p>A comparison of the outage probability of AF relay-assisted vehicular networks based on Equation (<a href="#FD23-electronics-08-00583" class="html-disp-formula">23</a>).</p> Full article ">Figure 8
<p>A comparison of the outage probability of AF relay-assisted vehicular networks based on Equation (<a href="#FD24-electronics-08-00583" class="html-disp-formula">24</a>).</p> Full article ">Figure 9
<p>The outage probability with different expressions vs. different outage thresholds, <math display="inline"><semantics> <msub> <mi>γ</mi> <mrow> <mi>t</mi> <mi>h</mi> </mrow> </msub> </semantics></math>, for <math display="inline"><semantics> <mrow> <msub> <mi>γ</mi> <mn>1</mn> </msub> <mo>=</mo> <msub> <mi>γ</mi> <mn>2</mn> </msub> <mo>,</mo> <mn>2</mn> <msub> <mi>γ</mi> <mn>2</mn> </msub> </mrow> </semantics></math>, and <math display="inline"><semantics> <mrow> <mn>3</mn> <msub> <mi>γ</mi> <mn>2</mn> </msub> </mrow> </semantics></math>.</p> Full article ">
27 pages, 11630 KiB  
Article
Improvement of Energy Efficiency and Control Performance of Cooling System Fan Applied to Industry 4.0 Data Center
by Jae-Sub Ko, Jun-Ho Huh and Jong-Chan Kim
Electronics 2019, 8(5), 582; https://doi.org/10.3390/electronics8050582 - 25 May 2019
Cited by 18 | Viewed by 5161
Abstract
This paper proposes a control method to improve the energy efficiency and performance of cooling fans used for cooling. In Industry 4.0, a large number of digital data are used, and a large number of data centers are created to handle these data. [...] Read more.
This paper proposes a control method to improve the energy efficiency and performance of cooling fans used for cooling. In Industry 4.0, a large number of digital data are used, and a large number of data centers are created to handle these data. These data centers consist of information technology (IT) equipment, power systems, and cooling systems. The cooling system is essential to prevent failure and malfunction of the IT equipment, which consumes a considerable amount of energy. This paper proposes a method to reduce the energy used in such cooling systems and to improve the temperature control performance. This paper proposes an fuzzy proportional integral(FPI) controller that controls the input value of the proportional integral(PI) controller by the fuzzy controller according to the operation state, a VFPI (Variable Fuzzy Proportional Integral) controller that adjusts the gain value of the fuzzy controller, and a variable fuzzy proportion integration-variable limit (VFPI-VL) controller that adjusts the limit value of the fuzzy controller’s output value. These controllers control the fan applied to the cooling system and compare the energy consumed and temperature control performance. When the PI controller consumes 100% of the power consumed, the FPI is 50.5%, the VFPI controller is 44.3%, and the VFPI-VL is 32.6%. The power consumption is greatly reduced. In addition, the VFPI-VL controller is the lowest in temperature variation, which improves the energy efficiency and performance of the cooling system using a fan. The methods presented in this paper can not only be applied to fans for cooling, but also to variable speed systems for various purposes and improvement of performance and efficiency can be expected. Full article
(This article belongs to the Section Artificial Intelligence)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Principle of the dehumidifier.</p> Full article ">Figure 2
<p>Principle of thermoelectric element.</p> Full article ">Figure 3
<p>Thermoelectric cooling system temperature characteristics.</p> Full article ">Figure 4
<p>Temperature characteristic of thermoelectric device. (<b>a</b>)heat absorption amount (<math display="inline"><semantics> <mrow> <msub> <mi mathvariant="normal">Q</mi> <mi mathvariant="normal">c</mi> </msub> </mrow> </semantics></math>); (<b>b</b>) Coefficient of performance (COP).</p> Full article ">Figure 5
<p>PI controller with anti-wind-up.</p> Full article ">Figure 6
<p>Membership function for the input value (error(e) and changing error(ce)): (<b>a</b>) The error (e) membership function, (<b>b</b>) the changing error (ce) membership function, and (<b>c</b>) output membership function.</p> Full article ">Figure 6 Cont.
<p>Membership function for the input value (error(e) and changing error(ce)): (<b>a</b>) The error (e) membership function, (<b>b</b>) the changing error (ce) membership function, and (<b>c</b>) output membership function.</p> Full article ">Figure 7
<p>The general structure of the fuzzy controller.</p> Full article ">Figure 8
<p>Structure of FPI controller.</p> Full article ">Figure 9
<p>Structure of variable gain fuzzy proportional integration (VFPI) controller.</p> Full article ">Figure 10
<p>Structure of variable gain fuzzy proportional integration with variable limit (VFPI-VL) controller.</p> Full article ">Figure 11
<p>Flowchart for control.</p> Full article ">Figure 12
<p>The system configuration diagram.</p> Full article ">Figure 13
<p>The temperature change according to the cooling on the hot side: (<b>a</b>) not cooled; (<b>b</b>) cooling.</p> Full article ">Figure 14
<p>Ambient temperature conditions for performance testing: (<b>a</b>) PI, (<b>b</b>) FPI, (<b>c</b>) VFPI with fixed limit, and (<b>d</b>) VFPI with variable limit.</p> Full article ">Figure 14 Cont.
<p>Ambient temperature conditions for performance testing: (<b>a</b>) PI, (<b>b</b>) FPI, (<b>c</b>) VFPI with fixed limit, and (<b>d</b>) VFPI with variable limit.</p> Full article ">Figure 15
<p>Temperature control by PI controller: (<b>a</b>) temperature and (<b>b</b>) Pulse width modulation (PWM) signal.</p> Full article ">Figure 16
<p>Input voltage, current, and power consumption (PI controller): (<b>a</b>) current and voltage and (<b>b</b>) power.</p> Full article ">Figure 17
<p>Temperature control by FPI controller: (<b>a</b>) temperature and (<b>b</b>) PWM signal.</p> Full article ">Figure 18
<p>Input voltage, current, and power consumption (FPI controller): (<b>a</b>) current and voltage and (<b>b</b>) power.</p> Full article ">Figure 18 Cont.
<p>Input voltage, current, and power consumption (FPI controller): (<b>a</b>) current and voltage and (<b>b</b>) power.</p> Full article ">Figure 19
<p>Temperature control by VFPI controller: (<b>a</b>) temperature and (<b>b</b>) PWM signal.</p> Full article ">Figure 19 Cont.
<p>Temperature control by VFPI controller: (<b>a</b>) temperature and (<b>b</b>) PWM signal.</p> Full article ">Figure 20
<p>Input voltage, current, and power consumption (VFPI): (<b>a</b>) current and voltage and (<b>b</b>) power.</p> Full article ">Figure 21
<p>Temperature control by VFPI-VL controller: (<b>a</b>) setting temperature difference and current temperature difference, (<b>b</b>) gain and variable limit value of fuzzy controller, and (<b>c</b>) PWM signal.</p> Full article ">Figure 21 Cont.
<p>Temperature control by VFPI-VL controller: (<b>a</b>) setting temperature difference and current temperature difference, (<b>b</b>) gain and variable limit value of fuzzy controller, and (<b>c</b>) PWM signal.</p> Full article ">Figure 22
<p>Input voltage, current, and power consumption (VFPI-VL): (<b>a</b>) current and voltage and (<b>b</b>) power.</p> Full article ">
12 pages, 6074 KiB  
Article
Four-State Coupled-Line Resonator for Chipless RFID Tags Application
by Wazie M. Abdulkawi and Abdel-Fattah A. Sheta
Electronics 2019, 8(5), 581; https://doi.org/10.3390/electronics8050581 - 25 May 2019
Cited by 17 | Viewed by 3795
Abstract
A novel quad-state coupled-line microstrip resonator is proposed for compact chipless radio frequency identification (RFID) tags. The proposed resonator can be reconfigured to present one of four possible states: 00, 01, 10, and 11, representing, no resonance, resonance at f2, resonance [...] Read more.
A novel quad-state coupled-line microstrip resonator is proposed for compact chipless radio frequency identification (RFID) tags. The proposed resonator can be reconfigured to present one of four possible states: 00, 01, 10, and 11, representing, no resonance, resonance at f2, resonance at f1, and resonance at both f1 and f2, respectively. The frequency span between f2 and f1 can be easily controlled, thereby reducing the required spectrum. Moreover, the proposed technique allows the storage of a large amount of data in a compact size to reduce the cost per bit. A multi-resonator prototype consisting of six resonators is designed, analyzed, and experimentally characterized. This prototype is implemented on the RT Duroid 5880 substrate with a dielectric constant of 2.2, loss tangent of 0.0009, and thickness of 0.79 mm. The designed configuration can be reconfigured for 46 codes. Two complete the RFID tags, including the six resonators and two orthogonally polarized transmitting and receiving antennas, are implemented and tested. The first tag code is designed for all ones, 111111111111, and the second tag is designed as 101010101010 code. Experimental results show good agreement with the simulation. Full article
(This article belongs to the Section Microwave and Wireless Communications)
►▼ Show Figures

Figure 1

Figure 1
<p>Simple block diagram of retransmission chipless radio frequency identification (RFID) tag.</p> Full article ">Figure 2
<p>Coupled resonator (<b>a</b>) geometry and (<b>b</b>) equivalent circuit.</p> Full article ">Figure 3
<p>Coupled bent resonator.</p> Full article ">Figure 4
<p>Geometry of the proposed resonator.</p> Full article ">Figure 5
<p>Four possible states for the single proposed resonator: (<b>a</b>) S<sub>1</sub> (no frequency), (<b>b</b>) S<sub>2</sub> (<span class="html-italic">f</span><sub>2</sub> only), (<b>c</b>) S<sub>3</sub> (<span class="html-italic">f</span><sub>1</sub> only), and (<b>d</b>) S<sub>4</sub> (<span class="html-italic">f</span><sub>1</sub> and <span class="html-italic">f</span><sub>2</sub>).</p> Full article ">Figure 6
<p>S<sub>21</sub> (<b>a</b>) and S<sub>11</sub> (<b>b</b>) responses for the single quad-state microstrip resonator.</p> Full article ">Figure 7
<p>Six quad-state resonators.</p> Full article ">Figure 8
<p>Simulated response of the proposed resonator tag with two different IDs: 111111111111 and 000000000000; (<b>a</b>) S<sub>21</sub> and (<b>b</b>) S<sub>11</sub>.</p> Full article ">Figure 9
<p>Simulated results of the proposed tag with three different codes: 111111111111, 101010101010, and 010101010101; (<b>a</b>) S<sub>21</sub> and (<b>b</b>) S<sub>11</sub>.</p> Full article ">Figure 10
<p>Fabricated codes: (<b>a</b>) 111111111111, (<b>b</b>) 101010101010, and (<b>c</b>) 010101010101.</p> Full article ">Figure 11
<p>Measured and simulated S<sub>21</sub> for code (111111111111).</p> Full article ">Figure 12
<p>Measured and simulated S<sub>21</sub> for code 101010101010.</p> Full article ">Figure 13
<p>Measured and simulated S<sub>21</sub> for code 010101010101.</p> Full article ">Figure 14
<p>UWB antenna (<b>a</b>) geometry and (<b>b</b>) fabricated circuit.</p> Full article ">Figure 15
<p>UWB antenna results (<b>a</b>) simulated and measured S<sub>11</sub> (<b>b</b>) simulated 3D radiation pattern.</p> Full article ">Figure 16
<p>Integrated chipless RFID tag prototype.</p> Full article ">Figure 17
<p>Measurement setup.</p> Full article ">Figure 18
<p>Measured and simulated S<sub>21</sub> of code 111111111111.</p> Full article ">Figure 19
<p>Measured and simulated S<sub>21</sub> of code 101010101010.</p> Full article ">
15 pages, 7351 KiB  
Article
Compact Printable Inverted-M Shaped Chipless RFID Tag Using Dual-Polarized Excitation
by Wazie M. Abdulkawi, Abdel-Fattah A. Sheta, Khaled Issa and Saleh A. Alshebeili
Electronics 2019, 8(5), 580; https://doi.org/10.3390/electronics8050580 - 25 May 2019
Cited by 21 | Viewed by 4386
Abstract
A novel and compact dual-polarized (DP) chipless radio-frequency identification (RFID) tag is presented in this paper. This tag can read both vertical and horizontal orientations within its frequency band, which improves the robustness and detection capability of the RFID system. The proposed tag [...] Read more.
A novel and compact dual-polarized (DP) chipless radio-frequency identification (RFID) tag is presented in this paper. This tag can read both vertical and horizontal orientations within its frequency band, which improves the robustness and detection capability of the RFID system. The proposed tag makes use of the slot length variation encoding technique to improve the encoding capacity. This technique can duplicate the encoding capacity, thereby reducing the overall tag size by almost 50%. In particular, the proposed tag has an encoding capacity of 20 bits in the 3–8 GHz frequency band and achieves data density of around 15.15 bits/cm2. Three prototypes are fabricated and tested outside an anechoic chamber. Furthermore, one tag is tested at different distances (10 cm, 30 cm, and 60 cm) from the reader and the measured results are compared. The simulated and measured results are in reasonable agreement, with acceptable shifts at some frequencies due to fabrication and experimental errors. Full article
(This article belongs to the Special Issue Advanced RFID Technology and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Geometry of the tag.</p> Full article ">Figure 2
<p>Ten inverted-M chipless RFID tags (<b>a</b>) active mode and (<b>b</b>) passive mode.</p> Full article ">Figure 3
<p>Simulation setup for the proposed tag.</p> Full article ">Figure 4
<p>Designed tag with two different bit combinations: (<b>a</b>) “0101010101” and (<b>b</b>) “1010101010”.</p> Full article ">Figure 5
<p>Amplitude reflection coefficients of the vertical (VP) and horizontal (HP) polarizations for states “1111111111” and “0000000000”.</p> Full article ">Figure 6
<p>Vertically (VP) and horizontally (HP) polarized S<sub>11</sub> of the proposed tag with state “1010101010”.</p> Full article ">Figure 7
<p>Vertically (VP) and horizontally (HP) polarized S<sub>11</sub> for code “0101010101”.</p> Full article ">Figure 8
<p>Amplitude difference between vertically and horizontally polarized S<sub>11</sub> for code “1111111111”.</p> Full article ">Figure 9
<p>Amplitude difference between vertically and horizontally polarized S<sub>11</sub> for code “1010101010”.</p> Full article ">Figure 10
<p>Amplitude difference between vertically and horizontally polarized S<sub>11</sub> for code “0101010101”.</p> Full article ">Figure 11
<p>Two simulated S<sub>11</sub> for the proposed tag with length variations.</p> Full article ">Figure 12
<p>Proposed tag with nine unit cells (ID: 11111111111111111111).</p> Full article ">Figure 13
<p>Fabricated images of the proposed tag with two different bit combinations: (<b>a</b>) 11001100110011001100 and (<b>b</b>) 00110011001100110011.</p> Full article ">Figure 14
<p>Measurement setup outside the anechoic chamber.</p> Full article ">Figure 15
<p>Simulated and measured S<sub>11</sub> under horizontal polarization for 10 cm distance from the RFID reader: (<b>a</b>) ID: 11111111111111111111, (<b>b</b>) ID: 11001100110011001100, and (<b>c</b>) ID: 00110011001100110011.</p> Full article ">Figure 16
<p>Simulated and measured S<sub>11</sub> under vertical polarization for 10 cm distance from the RFID reader: (<b>a</b>) ID: 11111111111111111111, (<b>b</b>) ID: 11001100110011001100, and (<b>c</b>) ID: 00110011001100110011.</p> Full article ">Figure 17
<p>Measured responses of the nine unit cells at different distances.</p> Full article ">
19 pages, 4365 KiB  
Article
The Tabu_Genetic Algorithm: A Novel Method for Hyper-Parameter Optimization of Learning Algorithms
by Baosu Guo, Jingwen Hu, Wenwen Wu, Qingjin Peng and Fenghe Wu
Electronics 2019, 8(5), 579; https://doi.org/10.3390/electronics8050579 - 25 May 2019
Cited by 38 | Viewed by 5109
Abstract
Machine learning algorithms have been widely used to deal with a variety of practical problems such as computer vision and speech processing. But the performance of machine learning algorithms is primarily affected by their hyper-parameters, as without good hyper-parameter values the performance of [...] Read more.
Machine learning algorithms have been widely used to deal with a variety of practical problems such as computer vision and speech processing. But the performance of machine learning algorithms is primarily affected by their hyper-parameters, as without good hyper-parameter values the performance of these algorithms will be very poor. Unfortunately, for complex machine learning models like deep neural networks, it is very difficult to determine their hyper-parameters. Therefore, it is of great significance to develop an efficient algorithm for hyper-parameter automatic optimization. In this paper, a novel hyper-parameter optimization methodology is presented to combine the advantages of a Genetic Algorithm and Tabu Search to achieve the efficient search for hyper-parameters of learning algorithms. This method is defined as the Tabu_Genetic Algorithm. In order to verify the performance of the proposed algorithm, two sets of contrast experiments are conducted. The Tabu_Genetic Algorithm and other four methods are simultaneously used to search for good values of hyper-parameters of deep convolutional neural networks. Experimental results show that, compared to Random Search and Bayesian optimization methods, the proposed Tabu_Genetic Algorithm finds a better model in less time. Whether in a low-dimensional or high-dimensional space, the Tabu_Genetic Algorithm has better search capabilities as an effective method for finding the hyper-parameters of learning algorithms. The presented method in this paper provides a new solution for solving the hyper-parameters optimization problem of complex machine learning models, which will provide machine learning algorithms with better performance when solving practical problems. Full article
(This article belongs to the Section Artificial Intelligence)
►▼ Show Figures

Figure 1

Figure 1
<p>A simple GA workflow.</p> Full article ">Figure 2
<p>Tabu_GA workflow.</p> Full article ">Figure 3
<p>A single point crossover operation.</p> Full article ">Figure 4
<p>Structure of the convolutional neural network for experiments on the Flower-5 dataset.</p> Full article ">Figure 5
<p>MNIST handwriting data set.</p> Full article ">Figure 6
<p>Average verification error curves for five methods in experiments on the MNIST dataset.</p> Full article ">Figure 7
<p>Flower-5 Dataset.</p> Full article ">Figure 8
<p>Average verification error curves for five methods in experiments on the Flower-5 dataset.</p> Full article ">Figure 9
<p>Average verification error curves for GA, TS, and Tabu_GA in additional experiments on the MNIST dataset.</p> Full article ">Figure 10
<p>Average verification error curves for GA, TS, and Tabu_GA in additional experiments on the Flower-5 dataset.</p> Full article ">
23 pages, 924 KiB  
Article
Time Series Analysis to Predict End-to-End Quality of Wireless Community Networks
by Pere Millan, Carles Aliagas, Carlos Molina, Emmanouil Dimogerontakis and Roc Meseguer
Electronics 2019, 8(5), 578; https://doi.org/10.3390/electronics8050578 - 25 May 2019
Cited by 5 | Viewed by 4490
Abstract
Community Networks have been around us for decades being initially deployed in the USA and Europe. They were designed by individuals to provide open and free “do it yourself” Internet access to other individuals in the same community and geographic area. In recent [...] Read more.
Community Networks have been around us for decades being initially deployed in the USA and Europe. They were designed by individuals to provide open and free “do it yourself” Internet access to other individuals in the same community and geographic area. In recent years, they have evolved as a viable solution to provide Internet access in developing countries and rural areas. Their social impact is measurable, as the community is provided with the right and opportunity of communication. Community networks combine wired and wireless links, and the nature of the wireless medium is unreliable. This poses several challenges to the routing protocol. For instance, Link-State routing protocols deal with End-to-End Quality tracking to select paths that maximize the delivery rate and minimize traffic congestion. In this work, we focused on End-to-End Quality prediction by means of time-series analysis to foresee which paths are more likely to change their quality. We show that it is possible to accurately predict End-to-End Quality with a small Mean Absolute Error in the routing layer of large-scale, distributed, and decentralized networks. In particular, we analyzed the path ETX behavior and properties to better identify the best prediction algorithm. We also analyzed the End-to-End Quality prediction accuracy some steps ahead in the future, as well as its dependency on the hour of the day. Besides, we quantified the computational cost of the prediction. Finally, we evaluated the impact of the usage for routing of our approach versus a simplified OLSR (ETX + Dijkstra) on an overloaded network. Full article
(This article belongs to the Section Networks)
►▼ Show Figures

Figure 1

Figure 1
<p>Frequency of the OLSR path hops.</p> Full article ">Figure 2
<p>Temporal evolution of the number of paths.</p> Full article ">Figure 3
<p>Persistence of paths.</p> Full article ">Figure 4
<p>Temporal evolution of the average ETX of paths (working days).</p> Full article ">Figure 5
<p>Temporal evolution of the average ETX of paths (non-working days).</p> Full article ">Figure 6
<p>Frequency of path lengths based on hop count and average ETX.</p> Full article ">Figure 7
<p>Deviation between average ETX and hops normalized by the corresponding number of hops.</p> Full article ">Figure 8
<p>Histogram of deviation between average ETX and number of hops.</p> Full article ">Figure 9
<p>Average Mean Absolute Error (MAE) of the EtEQ predictions.</p> Full article ">Figure 10
<p>Mean Absolute Error (MAE) of the EtEQ predictions as box plot.</p> Full article ">Figure 11
<p>Distribution of the RBR Mean Absolute Error (MAE) as box plot.</p> Full article ">Figure 12
<p>Histogram of the RBR Mean Absolute Error (MAE).</p> Full article ">Figure 13
<p>Evolution of the average ETX, the average prediction and the average absolute error.</p> Full article ">Figure 14
<p>Evolution of the average ETX, the average prediction and the average absolute error for: (<b>top</b>) 12:00–00:00 and (<b>bottom</b>) 00:00–12:00.</p> Full article ">Figure 15
<p>RBR average Mean Absolute Error (MAE) of the EtEQ predictions.</p> Full article ">Figure 16
<p>RBR Mean Absolute Error of EtEQ predictions, depicted as a box plot.</p> Full article ">Figure 17
<p>Execution time for one prediction with a preload model. One step means just the time of the next prediction. Ten steps means the average of ten subsequent predictions.</p> Full article ">Figure 18
<p>Execution time for Training WEKA model: mean is 14.68 s, maximum is 21.79 s and minimum is 5.31 s.</p> Full article ">Figure 19
<p>CPU utilization for Training WEKA model every 5 min in a period of 24 h: mean CPU utilization is about 3.78%.</p> Full article ">Figure 20
<p>Energy consumption in watts for Training WEKA model: mean is 11.99 W, idle is 10.14 W, and difference is 1.86 W.</p> Full article ">Figure 21
<p>Temperature variation for Training WEKA model every 5 min in a period of 24 h: mean is 40.84 <math display="inline"><semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics></math>C, maximum is 47 <math display="inline"><semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics></math>C and minimum is 37 <math display="inline"><semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics></math>C.</p> Full article ">Figure 22
<p>Histogram of the ETX of links.</p> Full article ">Figure 23
<p>Histogram of end-to-end delay.</p> Full article ">
17 pages, 3098 KiB  
Article
Risk-Constrained Stochastic Scheduling of a Grid-Connected Hybrid Microgrid with Variable Wind Power Generation
by Mostafa Vahedipour-Dahraie, Homa Rashidizadeh-Kermani and Amjad Anvari-Moghaddam
Electronics 2019, 8(5), 577; https://doi.org/10.3390/electronics8050577 - 25 May 2019
Cited by 11 | Viewed by 2865
Abstract
This paper presents a risk-constrained scheduling optimization model for a grid-connected hybrid microgrid including demand response (DR), electric vehicles (EVs), variable wind power generation and dispatchable generation units. The proposed model determines optimal scheduling of dispatchable units, interactions with the main grid as [...] Read more.
This paper presents a risk-constrained scheduling optimization model for a grid-connected hybrid microgrid including demand response (DR), electric vehicles (EVs), variable wind power generation and dispatchable generation units. The proposed model determines optimal scheduling of dispatchable units, interactions with the main grid as well as adjustable responsive loads and EVs demand to maximize the expected microgrid operator’s profit under different scenarios. The uncertainties of day-ahead (DA) market prices, wind power production and demands of customers and EVs are considered in this study. To address these uncertainties, conditional value-at-risk (CVaR) as a risk measurement tool is added to the optimization model to evaluate the risk of profit loss and to indicate decision attitudes in different conditions. The proposed method is finally applied to a typical hybrid microgrid with flexible demand-side resources and its applicability and effectives are verified over different working conditions with uncertainties. Full article
(This article belongs to the Special Issue Active Regional Energy Systems and Microgrids)
►▼ Show Figures

Figure 1

Figure 1
<p>Flowchart of the proposed algorithm for solving the optimal scheduling model.</p> Full article ">Figure 2
<p>Schematic diagram of the microgrid being studied.</p> Full article ">Figure 3
<p>The hourly forecasted values of (<b>a</b>) customers’ loads, charging demand of electric vehicles (EVs) and wind power and (<b>b</b>) day-ahead (DA) electricity price.</p> Full article ">Figure 4
<p>Operator’s expected profit versus conditional value-at-risk (CVaR) in different wind penetration levels: (<b>a</b>) Wind penetration level = 0%; (<b>b</b>) Wind penetration level = 50%, and (<b>c</b>) Wind penetration level = 100%.</p> Full article ">Figure 4 Cont.
<p>Operator’s expected profit versus conditional value-at-risk (CVaR) in different wind penetration levels: (<b>a</b>) Wind penetration level = 0%; (<b>b</b>) Wind penetration level = 50%, and (<b>c</b>) Wind penetration level = 100%.</p> Full article ">Figure 5
<p>Impact of wind power penetration level at different values of <span class="html-italic">β</span> on (<b>a</b>) expected profit, (<b>b</b>) cost of generation, and (<b>c</b>) CVaR.</p> Full article ">Figure 5 Cont.
<p>Impact of wind power penetration level at different values of <span class="html-italic">β</span> on (<b>a</b>) expected profit, (<b>b</b>) cost of generation, and (<b>c</b>) CVaR.</p> Full article ">Figure 6
<p>Cost of trading energy versus wind penetration level (<b>a</b>) buying from main grid; (<b>b</b>) selling to the main grid.</p> Full article ">Figure 7
<p>Cost of trading energy versus risk aversion parameter (<b>a</b>) buying from the main grid, (<b>b</b>) selling to the main grid.</p> Full article ">Figure 8
<p>Hourly energy trading between the microgrid and the main grid in different penetration levels of wind power, (<b>a</b>) penetration level = 0%, (<b>b</b>) penetration level = 50%, and (<b>c</b>) penetration level = 100%.</p> Full article ">Figure 8 Cont.
<p>Hourly energy trading between the microgrid and the main grid in different penetration levels of wind power, (<b>a</b>) penetration level = 0%, (<b>b</b>) penetration level = 50%, and (<b>c</b>) penetration level = 100%.</p> Full article ">
19 pages, 9525 KiB  
Article
Completing Explorer Games with a Deep Reinforcement Learning Framework Based on Behavior Angle Navigation
by Shixun You, Ming Diao and Lipeng Gao
Electronics 2019, 8(5), 576; https://doi.org/10.3390/electronics8050576 - 25 May 2019
Cited by 4 | Viewed by 3683
Abstract
In cognitive electronic warfare, when a typical combat vehicle, such as an unmanned combat air vehicle (UCAV), uses radar sensors to explore an unknown space, the target-searching fails due to an inefficient servoing/tracking system. Thus, to solve this problem, we developed an autonomous [...] Read more.
In cognitive electronic warfare, when a typical combat vehicle, such as an unmanned combat air vehicle (UCAV), uses radar sensors to explore an unknown space, the target-searching fails due to an inefficient servoing/tracking system. Thus, to solve this problem, we developed an autonomous reasoning search method that can generate efficient decision-making actions and guide the UCAV as early as possible to the target area. For high-dimensional continuous action space, the UCAV’s maneuvering strategies are subject to certain physical constraints. We first record the path histories of the UCAV as a sample set of supervised experiments and then construct a grid cell network using long short-term memory (LSTM) to generate a new displacement prediction to replace the target location estimation. Finally, we enable a variety of continuous-control-based deep reinforcement learning algorithms to output optimal/sub-optimal decision-making actions. All these tasks are performed in a three-dimensional target-searching simulator, i.e., the Explorer game. Please note that we use the behavior angle (BHA) for the first time as the main factor of the reward-shaping of the deep reinforcement learning framework and successfully make the trained UCAV achieve a 99.96% target destruction rate, i.e., the game win rate, in a 0.1 s operating cycle. Full article
(This article belongs to the Section Systems & Control Engineering)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) displays the interface of the Explorer game, with a total pixel size of <math display="inline"><semantics> <mrow> <mn>600</mn> <mo>×</mo> <mn>1050</mn> </mrow> </semantics></math>; the <math display="inline"><semantics> <mrow> <mn>600</mn> <mo>×</mo> <mn>600</mn> </mrow> </semantics></math> panel on the left side displays the map from the global perspective, and the <math display="inline"><semantics> <mrow> <mn>600</mn> <mo>×</mo> <mn>450</mn> </mrow> </semantics></math> panel on the right side displays the status information of the UCAV and the observation station. (<b>b</b>) shows the maximum detection range of the UCAV (reconnaissance perspective). The ratio with respect to the real environment is 1:25, i.e., 1 pixel corresponds to 25 m.</p> Full article ">Figure 2
<p>Grid network architecture in the supervised learning experiment. The recurrent layer is an LSTM with 128 hidden units. <math display="inline"><semantics> <mi mathvariant="bold-italic">c</mi> </semantics></math> and <math display="inline"><semantics> <mi mathvariant="bold-italic">h</mi> </semantics></math> represent the place cell activations and head-direction cell activations, respectively.</p> Full article ">Figure 3
<p>Schematic diagram of the behavior angle of UCAV in motion.</p> Full article ">Figure 4
<p>The behavior incentives provided by different reward functions: (<b>a</b>–<b>c</b>) correspond to <math display="inline"><semantics> <msub> <mi>R</mi> <mi>t</mi> </msub> </semantics></math>, <math display="inline"><semantics> <msubsup> <mi>R</mi> <mi>t</mi> <mo>′</mo> </msubsup> </semantics></math>, and <math display="inline"><semantics> <msubsup> <mi>R</mi> <mi>t</mi> <mrow> <mo>″</mo> </mrow> </msubsup> </semantics></math>, respectively. The thickness of the red arrow indicates the proportion of information dissemination.</p> Full article ">Figure 5
<p>The actor-critic (AC) framework.</p> Full article ">Figure 6
<p>DRL-based cognitive system for Explorer. Please note that the <math display="inline"><semantics> <msub> <mi>S</mi> <mi>t</mi> </msub> </semantics></math> is the cognitive state of the UCAV, and the <math display="inline"><semantics> <msub> <mi>R</mi> <mi>t</mi> </msub> </semantics></math> is the real reward induced by the real state of the Explorer game.</p> Full article ">Figure 7
<p>Evaluating the convergence performance of the DRL framework in Explorer. The twelve subfigures from left to right correspond to the results for game levels 0 to 3, and from top to bottom correspond to the reward functions <math display="inline"><semantics> <msub> <mi>R</mi> <mi>t</mi> </msub> </semantics></math>, <math display="inline"><semantics> <msubsup> <mi>R</mi> <mi>t</mi> <mo>′</mo> </msubsup> </semantics></math>, and <math display="inline"><semantics> <msubsup> <mi>R</mi> <mi>t</mi> <mrow> <mo>″</mo> </mrow> </msubsup> </semantics></math>, respectively.</p> Full article ">Figure 8
<p>Investigating the potential behavior policies learned by the UCAV in Explorer. The twelve subfigures from left to right correspond to the results for game levels 0 to 3, and from top to bottom correspond to the reward functions <math display="inline"><semantics> <msub> <mi>R</mi> <mi>t</mi> </msub> </semantics></math>, <math display="inline"><semantics> <msubsup> <mi>R</mi> <mi>t</mi> <mo>′</mo> </msubsup> </semantics></math>, and <math display="inline"><semantics> <msubsup> <mi>R</mi> <mi>t</mi> <mrow> <mo>″</mo> </mrow> </msubsup> </semantics></math>, respectively.</p> Full article ">
15 pages, 4241 KiB  
Review
Review of the Recent Progress on GaN-Based Vertical Power Schottky Barrier Diodes (SBDs)
by Yue Sun, Xuanwu Kang, Yingkui Zheng, Jiang Lu, Xiaoli Tian, Ke Wei, Hao Wu, Wenbo Wang, Xinyu Liu and Guoqi Zhang
Electronics 2019, 8(5), 575; https://doi.org/10.3390/electronics8050575 - 24 May 2019
Cited by 78 | Viewed by 12722
Abstract
Gallium nitride (GaN)-based vertical power Schottky barrier diode (SBD) has demonstrated outstanding features in high-frequency and high-power applications. This paper reviews recent progress on GaN-based vertical power SBDs, including the following sections. First, the benchmark for GaN vertical SBDs with different substrates (Si, [...] Read more.
Gallium nitride (GaN)-based vertical power Schottky barrier diode (SBD) has demonstrated outstanding features in high-frequency and high-power applications. This paper reviews recent progress on GaN-based vertical power SBDs, including the following sections. First, the benchmark for GaN vertical SBDs with different substrates (Si, sapphire, and GaN) are presented. Then, the latest progress in the edge terminal techniques are discussed. Finally, a typical fabrication flow of vertical GaN SBDs is also illustrated briefly. Full article
(This article belongs to the Section Semiconductor Devices)
►▼ Show Figures

Figure 1

Figure 1
<p>Gallium nitride Schottky barrier diode (GaN SBD) structure and current flowing directions: (<b>a</b>) quasi-vertical and (<b>b</b>) fully-vertical.</p> Full article ">Figure 2
<p>Benchmarks of the R<sub>ON,sp</sub> vs. breakdown voltage (BV) of vertical SBDs with GaN on Si, sapphire, and GaN substrates.</p> Full article ">Figure 3
<p>(<b>a</b>) Basic SBD structure and (<b>b</b>) electric field distribution.</p> Full article ">Figure 4
<p>Electric field distribution of a typical vertical SBD under reverse bias.</p> Full article ">Figure 5
<p>(<b>a</b>) Cross-section of GaN SBDs with p-guard ring edge termination structure and (<b>b</b>) E-field distribution under reverse bias.</p> Full article ">Figure 6
<p>Schematic cross-section of a field plate with GaN SBDs: (<b>a</b>) metal field plate, (<b>b</b>) resistive field plate, (<b>c</b>) floating field plate, and (<b>d</b>) E-field distribution under reverse bias.</p> Full article ">Figure 6 Cont.
<p>Schematic cross-section of a field plate with GaN SBDs: (<b>a</b>) metal field plate, (<b>b</b>) resistive field plate, (<b>c</b>) floating field plate, and (<b>d</b>) E-field distribution under reverse bias.</p> Full article ">Figure 7
<p>Schematic cross-section of vertical GaN TMBS: (<b>a</b>) typical structure, (<b>b</b>) FR-TMBS structure, and (<b>c</b>) simulated E-field distribution of FR-TMBS at −1000 V.</p> Full article ">Figure 8
<p>(<b>a</b>) Schematic cross-sections of GaN reduced surface field (RESURF) junction barrier-controlled Schottky (JBS) structure and (<b>b</b>) comparison of simulated E-field distribution of typical SBD with RESURF JBS at −200 V.</p> Full article ">Figure 9
<p>(<b>a</b>) Schematic cross-section of the vertical GaN NT-SBD and (<b>b</b>) simulated leakage current distribution of NT-SBD at −300 V.</p> Full article ">Figure 10
<p>Benchmarks of the R<sub>ON,sp</sub> vs. BV of vertical SBDs with different edge termination techniques.</p> Full article ">Figure 11
<p>Vertical GaN SBD device process flowchart: (<b>a</b>) ohmic contact, (<b>b</b>) nickel electroplating, (<b>c</b>) laser lift-off, (<b>d</b>) sapphire removal, and (<b>e</b>) Schottky contact.</p> Full article ">
10 pages, 5352 KiB  
Article
A Doping-Less Tunnel Field-Effect Transistor with Si0.6Ge0.4 Heterojunction for the Improvement of the On–Off Current Ratio and Analog/RF Performance
by Tao Han, Hongxia Liu, Shupeng Chen, Shulong Wang and Wei Li
Electronics 2019, 8(5), 574; https://doi.org/10.3390/electronics8050574 - 24 May 2019
Cited by 8 | Viewed by 3670
Abstract
In this paper, a novel doping-less tunneling field-effect transistor with Si0.6Ge0.4 heterojunction (H-DLTFET) is proposed using TCAD simulation. Unlike conventional doping-less tunneling field-effect transistors (DLTFETs), in H-DLTFETs, germanium and Si0.6Ge0.4 are used as source and channel materials, [...] Read more.
In this paper, a novel doping-less tunneling field-effect transistor with Si0.6Ge0.4 heterojunction (H-DLTFET) is proposed using TCAD simulation. Unlike conventional doping-less tunneling field-effect transistors (DLTFETs), in H-DLTFETs, germanium and Si0.6Ge0.4 are used as source and channel materials, respectively, to provide higher carrier mobility and smaller tunneling barrier width. The energy band and charge carrier tunneling efficiency of the tunneling junction become steeper and higher as a result of the Si0.6Ge0.4 heterojunction. In addition, the effects of the source work function, gate oxide dielectric thickness, and germanium content on the performance of the H-DLTFET are analyzed systematically, and the below optimal device parameters are obtained. The simulation results show that the performance parameters of the H-DLTFET, such as the on-state current, on/off current ratio, output current, subthreshold swing, total gate capacitance, cutoff frequency, and gain bandwidth (GBW) product when Vd = 1 V and Vg = 2 V, are better than those of conventional silicon-based DLTFETs. Therefore, the H-DLTFET has better potential for use in ultra-low power devices. Full article
(This article belongs to the Section Semiconductor Devices)
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic diagrams of (<b>a</b>) a conventional tunneling field-effect transistor (DLTFET) and (<b>b</b>) the new proposed doping-less tunneling field-effect transistor with Si<sub>0.6</sub>Ge<sub>0.4</sub> heterojunction (H-DLTFET).</p> Full article ">Figure 2
<p>(<b>a</b>) Transfer characteristics; (<b>b</b>) transconductance of the DLTFET and the H-DLTFET.</p> Full article ">Figure 3
<p>Output characteristics: (<b>a</b>) the DLTFET and (<b>b</b>) the H-DLTFET at different gate voltages.</p> Full article ">Figure 4
<p>(<b>a</b>) Electronic band-to-band tunneling (BTBT) generation rate; (<b>b</b>) hole BTBT generation rate; (<b>c</b>) electric field distribution; and (<b>d</b>) total current density distribution of the H-DLTFET.</p> Full article ">Figure 5
<p>Schematic diagram of the energy band according to the on-state condition (<b>a</b>) in the middle of the channel from the top channel to the bottom channel and (<b>b</b>) on the channel surface from the source to the drain.</p> Full article ">Figure 6
<p>(<b>a</b>) Influence of T<sub>ox</sub> on the transfer characteristic; (<b>b</b>) on-state energy band at the surface of channel from the source to the drain.</p> Full article ">Figure 7
<p>Influence of X<sub>Ge</sub> on (<b>a</b>) the transfer characteristic and (<b>b</b>) the on-state energy band from the top channel to the bottom channel.</p> Full article ">Figure 8
<p>Influence of Ψs on (<b>a</b>) the transfer characteristic and (<b>b</b>) the on-state energy band from the top channel to the bottom channel.</p> Full article ">Figure 9
<p>C-V characteristics of (<b>a</b>) the DLTFET and (<b>b</b>) the H-DLTFET.</p> Full article ">Figure 10
<p>(<b>a</b>) Cutoff frequency and (<b>b</b>) the gain bandwidth product of the DLTFET and the H-DLTFET.</p> Full article ">
16 pages, 9767 KiB  
Article
Weak Signal Extraction from Lunar Penetrating Radar Channel 1 Data Based on Local Correlation
by Zhuo Jia, Sixin Liu, Ling Zhang, Bin Hu and Jianmin Zhang
Electronics 2019, 8(5), 573; https://doi.org/10.3390/electronics8050573 - 23 May 2019
Cited by 8 | Viewed by 3172
Abstract
Knowledge of the subsurface structure not only provides useful information on lunar geology, but it also can quantify the potential lunar resources for human beings. The dual-frequency lunar penetrating radar (LPR) aboard the Yutu rover offers a Special opportunity to understand the subsurface [...] Read more.
Knowledge of the subsurface structure not only provides useful information on lunar geology, but it also can quantify the potential lunar resources for human beings. The dual-frequency lunar penetrating radar (LPR) aboard the Yutu rover offers a Special opportunity to understand the subsurface structure to a depth of several hundreds of meters using a low-frequency channel (channel 1), as well as layer near-surface stratigraphic structure of the regolith using high-frequency observations (channel 2). The channel 1 data of the LPR has a very low signal-to-noise ratio. However, the extraction of weak signals from the data represents a problem worth exploring. In this article, we propose a weak signal extraction method in view of local correlation to analyze the LPR CH-1 data, to facilitate a study of the lunar regolith structure. First, we build a pre-processing workflow to increase the signal-to-noise ratio (SNR). Second, we apply the K-L transform to separate the horizontal signal and then use the seislet transform (ST) to reserve the continuous signal. Then, the local correlation map is calculated using the two denoising results and a time–space dependent weighting operator is constructed to suppress the noise residuals. The weak signal after noise suppression may provide a new reference for subsequent data interpretation. Finally, in combination with the regional geology and previous research, we provide some speculative interpretations of the LPR CH-1 data. Full article
(This article belongs to the Special Issue Advanced Technology Related to Radar Signal, Imaging, and Radar Cross-Section Measurement)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Yutu’s path on the Moon. The background image was taken by the descent camera on the Chang’E-3 (CE-3) lander. The red star shows the landing site. The inset line shows the path (purple line and red line).</p> Full article ">Figure 2
<p>The antennas on Yutu Rover. (<b>a</b>) Yutu rover; (<b>b</b>) CH-1; (<b>c</b>) CH-2.</p> Full article ">Figure 3
<p>The flowchart of the CH-1 (lunar penetrating radar) LPR data preprocessing (left). The output is the LPR CH-1 data after preprocessing (right). N104–N207 denote the positions where the LPR was rebooted.</p> Full article ">Figure 4
<p>LPR CH-1 data with interpretation.</p> Full article ">Figure 5
<p>Demonstration of weighting operator.</p> Full article ">Figure 6
<p>Demonstration of weak signal extraction. (<b>a</b>) Signal; (<b>b</b>) Noisy signal; (<b>c</b>) Pre-denoised result: K-L transform; (<b>d</b>) Pre-denoised result: ST; (<b>e</b>) Local correlation map of (<b>c</b>,<b>d</b>); (<b>f</b>) Extracted signal using our method.</p> Full article ">Figure 7
<p>Demonstration of pre-denoised results. (<b>a</b>) K-L transform; (<b>b</b>) ST.</p> Full article ">Figure 8
<p>(<b>a</b>) Local correlation map and (<b>b</b>) final extracted signal.</p> Full article ">Figure 9
<p>Zoomed local correlation map and zoomed section of the final extracted signal. (<b>a</b>) Local correlation map (Near 290m) (<b>b</b>) Extracted signal (Near 290m) (<b>c</b>) Local correlation map (Near 450m) Local correlation map (<b>d</b>) Extracted signal (Near 450m).</p> Full article ">Figure 10
<p>Demonstration of the extracted weak signal (with depth marked). (<b>a</b>) Local correlation map (<b>b</b>) Extracted signal using our method.</p> Full article ">Figure 11
<p>Interpretation of the data.</p> Full article ">
18 pages, 4713 KiB  
Article
Design and Analysis of Three-Stage Amplifier for Driving pF-to-nF Capacitive Load Based on Local Q-Factor Control and Cascode Miller Compensation Techniques
by Qi Cheng, Weimin Li, Xian Tang and Jianping Guo
Electronics 2019, 8(5), 572; https://doi.org/10.3390/electronics8050572 - 23 May 2019
Cited by 13 | Viewed by 6614
Abstract
This paper presents a new frequency compensation approach for three-stage amplifiers driving a pF-to-nF capacitive load. Thanks to the cascode Miller compensation, the non-dominant complex pole frequency is extended effectively, and the physical size of the compensation capacitors is also reduced. A local [...] Read more.
This paper presents a new frequency compensation approach for three-stage amplifiers driving a pF-to-nF capacitive load. Thanks to the cascode Miller compensation, the non-dominant complex pole frequency is extended effectively, and the physical size of the compensation capacitors is also reduced. A local Q-factor control (LQC) loop is introduced to alter the Q-factor adaptively when loading capacitance CL varies significantly. This LQC loop decides how much damping current should be injected into the corresponding parasitic node to control the Q-factor of the complex-pole pair, which affects the frequency peak at the gain plot and the settling time of the proposed amplifier in the closed-loop step response. Additionally, a left-half-plane (LHP) zero is created to increase the phase margin and a feed-forward transconductance stage is paralleled to improve the slew rate (SR). Simulated in 0.13-µm CMOS technology, the amplifier is verified to handle a 4-pF-to-1.5-nF (375× drivability) capacitive load with at least 0.88-MHz gain-bandwidth (GBW) product and 42.3° phase margin (PM), while consuming 24.0-µW quiescent power at 1.0-V nominal supply voltage. Full article
(This article belongs to the Special Issue Low-Voltage Integrated Circuits Design and Application)
►▼ Show Figures

Figure 1

Figure 1
<p>Three-stage nested-Miller compensation (NMC) amplifier: (<b>a</b>) topology, and (<b>b</b>) the pole-zeros locus under 100× <span class="html-italic">C<sub>L</sub></span> variation (unscaled).</p> Full article ">Figure 2
<p>Three-stage damping factor control frequency compensation (DFCFC) amplifier: (<b>a</b>) topology, and (<b>b</b>) the pole-zeros locus under 100× <span class="html-italic">C<sub>L</sub></span> variation (unscaled).</p> Full article ">Figure 3
<p>Three-stage cascode Miller compensation with local impedance attenuation (CLIA) amplifier: (<b>a</b>) topology, and (<b>b</b>) the pole-zeros locus under 100× <span class="html-italic">C<sub>L</sub></span> variation (unscaled).</p> Full article ">Figure 4
<p>Equivalent diagram of the proposed three-stage cascode Miller-compensation with local <span class="html-italic">Q</span>-factor control (CLQC) amplifier.</p> Full article ">Figure 5
<p>The equivalent small-signal model of the proposed three-stage CLQC amplifier.</p> Full article ">Figure 6
<p>Pole locus of the proposed amplifier when the load capacitance varies from <span class="html-italic">C<sub>L</sub></span> to <span class="html-italic">C<sub>L</sub></span>/1000 (scaled).</p> Full article ">Figure 7
<p>Frequency responses of NMC, cascode compensation without and with Local <span class="html-italic">Q</span>-Factor Control (LQC).</p> Full article ">Figure 8
<p>The simplified schematic of the proposed three-stage CLQC amplifier.</p> Full article ">Figure 9
<p>Layout of the proposed amplifier circuit.</p> Full article ">Figure 10
<p>AC simulation results with various values of <span class="html-italic">C<sub>L</sub></span> ranging from 4 pF to 1.5 nF.</p> Full article ">Figure 11
<p>AC simulation results (<b>a</b>–<b>d</b>) with <span class="html-italic">C<sub>L</sub></span> (4 pF to 1.5 nF) when <span class="html-italic">V<sub>DD</sub></span> varies from 0.8 to 1.2 V.</p> Full article ">Figure 12
<p>Simulated 500 mV step responses at (<b>a</b>) <span class="html-italic">C<sub>L</sub></span> = 4 pF (<b>b</b>) <span class="html-italic">C<sub>L</sub></span> = 150 pF (<b>c</b>) <span class="html-italic">C<sub>L</sub></span> = 500 pF and (<b>d</b>) <span class="html-italic">C<sub>L</sub></span> = 1.5 nF.</p> Full article ">Figure 13
<p>Simulated results of the proposed amplifier (<b>a</b>) PSRR (<b>b</b>) CMRR and (<b>c</b>) output noise density.</p> Full article ">
18 pages, 3557 KiB  
Article
Characteristic Impedance Analysis of Medium-Voltage Underground Cables with Grounded Shields and Armors for Power Line Communication
by Hongshan Zhao, Weitao Zhang and Yan Wang
Electronics 2019, 8(5), 571; https://doi.org/10.3390/electronics8050571 - 23 May 2019
Cited by 7 | Viewed by 5053
Abstract
The characteristic impedance of a power line is an important parameter in power line communication (PLC) technologies. This parameter is helpful for understanding power line impedance characteristics and achieving impedance matching. In this study, we focused on the characteristic impedance matrices (CIMs) of [...] Read more.
The characteristic impedance of a power line is an important parameter in power line communication (PLC) technologies. This parameter is helpful for understanding power line impedance characteristics and achieving impedance matching. In this study, we focused on the characteristic impedance matrices (CIMs) of the medium-voltage (MV) cables. The calculation and characteristics of the CIMs were investigated with special consideration of the grounded shields and armors, which are often neglected in current research. The calculation results were validated through the experimental measurements. The results show that the MV underground cables with multiple grounding points have forward and backward CIMs, which are generally not equal unless the whole cable structure is longitudinally symmetrical. Then, the resonance phenomenon in the CIMs was analyzed. We found that the grounding of the shields and armors not only affected their own characteristic impedances but also those of the cores, and the resonance present in the CIMs should be of concern in the impedance matching of the PLC systems. Finally, the effects of the grounding resistances, cable lengths, grounding point numbers, and cable branch numbers on the CIMs of the MV underground cables were discussed through control experiments. Full article
(This article belongs to the Special Issue Advances of Power Line Communication (PLC))
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Structure of the medium-voltage underground cable. XPLE stands for cross-linked polyethylene, PVC stands for polyvinyl chloride; (<b>b</b>) Voltage and current definitions of the A-phase core, A phase shield, and the armor.</p> Full article ">Figure 2
<p>Structure of medium-voltage underground cables with shields and armors grounding at the ring main units (RMUs).</p> Full article ">Figure 3
<p>Measurement setup. The measurement of (<b>a</b>) the sending port’s access impedance matrix and (<b>b</b>) the receiving port’s access impedance matrix.</p> Full article ">Figure 4
<p>The validation of the forward characteristic impedances of medium-voltage underground cables with grounded shields and armor in example 1: (<b>a</b>) The real parts and (<b>b</b>) the imaginary parts of the characteristic impedances.</p> Full article ">Figure 5
<p>The validation of the backward characteristic impedances of medium-voltage underground cables with grounded shields and armor in example 1: (<b>a</b>) The real parts and (<b>b</b>) the imaginary parts of the characteristic impedances.</p> Full article ">Figure 6
<p>The characteristic impedances of medium-voltage underground cables with grounded shields and armor in example 2: (<b>a</b>) The numerical results and (<b>b</b>) the numerical results of the backward characteristic impedances.</p> Full article ">Figure 7
<p>The validation of the forward characteristic impedances of medium-voltage underground cables with grounded shields and armor in example 2: (<b>a</b>) The real parts and (<b>b</b>) the imaginary parts of the characteristic impedances.</p> Full article ">Figure 8
<p>The comparation of the forward characteristic impedances when the cables are lossy and lossless: (<b>a</b>) The real parts and (<b>b</b>) the imaginary parts of the characteristic impedances.</p> Full article ">Figure 9
<p>The forward characteristic impedances of medium-voltage underground cables with different grounding resistances: (<b>a</b>) The real parts and (<b>b</b>) the imaginary parts of the characteristic impedances.</p> Full article ">Figure 10
<p>The forward characteristic impedances of medium-voltage underground cables with different cable lengths: (<b>a</b>) The real parts and (<b>b</b>) the imaginary parts of the characteristic impedances.</p> Full article ">Figure 11
<p>The forward characteristic impedances of medium-voltage underground cables with different grounding point numbers: (<b>a</b>) The real parts and (<b>b</b>) the imaginary parts of the characteristic impedances.</p> Full article ">Figure 12
<p>The forward characteristic impedances of medium-voltage underground cables with different branch numbers: (<b>a</b>) The real parts and (<b>b</b>) the imaginary parts of the characteristic impedances.</p> Full article ">
19 pages, 1272 KiB  
Article
Emerging Zero-Standby Solutions for Miscellaneous Electric Loads and the Internet of Things
by Daniel L. Gerber, Alan Meier, Richard Liou and Robert Hosbach
Electronics 2019, 8(5), 570; https://doi.org/10.3390/electronics8050570 - 23 May 2019
Cited by 16 | Viewed by 7076
Abstract
Despite technical advances in efficiency, devices in standby continue to consume up to 16% of residential electricity. Finding practical, cost-effective reductions is difficult. While the per-unit power consumption has fallen, the number of units continuously drawing power continues to grow. This work reviews [...] Read more.
Despite technical advances in efficiency, devices in standby continue to consume up to 16% of residential electricity. Finding practical, cost-effective reductions is difficult. While the per-unit power consumption has fallen, the number of units continuously drawing power continues to grow. This work reviews a family of technologies that can eliminate standby consumption in many types of electrical plug loads. It also investigates several solutions in detail and develops prototypes. First, burst mode and sleep transistors are established as building blocks for zero-standby solutions. This work then studies the application of two types of wake-up signals. The first is from an optical transmission, and is applicable to remote-controlled devices with a line-of-sight activation, such as set-top boxes, ceiling fans, and motorized curtains. The second is from a wake-up radio, and is applicable to any wireless products. No single technology will address all standby power situations; however, these emerging solutions appear to have broad applicability to save standby energy in miscellaneous plug loads. Full article
(This article belongs to the Section Power Electronics)
►▼ Show Figures

Figure 1

Figure 1
<p>The distribution of a computer’s power modes with respect to time [<a href="#B12-electronics-08-00570" class="html-bibr">12</a>]. For an entry-level desktop with light use, standby modes account for 25% of the total energy consumption.</p> Full article ">Figure 2
<p>A portfolio of standby reduction techniques and solutions. The solutions discussed in this paper are shown in the bright green boxes with bold font.</p> Full article ">Figure 3
<p>Burst mode requires a supply capacitor and its necessary recharge logic. Converters without built-in burst mode can still be used in burst mode if they have an enable (EN In) pin.</p> Full article ">Figure 4
<p>A simple design for the burst mode logic. The resistor divider scales the supply capacitor voltage, which is compared to high and low references. The reference voltages can be generated by nano-watt band gap references or low dropout regulator (LDOs). The comparators drive an SR latch, which determines if the converter is enabled.</p> Full article ">Figure 5
<p>Oscilloscope waveforms with the supply capacitor voltage (yellow), low threshold comparator output (green), and high threshold comparator output (blue). These waveforms are from the PG02S2405A, loaded with a 4.7 kΩ resistor.</p> Full article ">Figure 6
<p>A footer switch connects the device ground to the supply ground. A wake-up signal is required to drive the gate of the footer switch.</p> Full article ">Figure 7
<p>A header switch topology. The wake-up signal activates an NMOS, which activates the PMOS header switch through a pull-up resistor.</p> Full article ">Figure 8
<p>A cascoded header switch topology with example transistor values for a 48 V DC supply. The biasing network functions to bias the gate of M2 at 10 V. The pull-up network contains a Zener diode to protect the gate of the header switch M3. The latch is not shown.</p> Full article ">Figure 9
<p>Proposed IR energy harvesting method for set-top boxes: (<b>a</b>) when the power button is pressed, a high power IR signal is transmitted to wake the device; and (<b>b</b>) once the device is awake, ordinary low power IR signals can be used for all other functions (e.g., changing the channel).</p> Full article ">Figure 10
<p>The receiver for a zero standby supply with an IR-based wakeup signal. The photodiode harvests IR energy from the IR signal, which drives the gate of the footer switch.</p> Full article ">Figure 11
<p>A prototype of the IR-based zero standby supply. The transmitter with four IR LEDs is shown on the left, the receiver with 12 dark photodiodes and the device (LCD screen) are shown on the right.</p> Full article ">Figure 12
<p>Laser standby solution with a four-stage Dickson charge pump attached to an NMOS footer switch M1.</p> Full article ">Figure 13
<p>The prototypes of the laser-based zero standby supply. (<b>a</b>) The charge pump prototype in a battery-powered lamp. The receiver is shown on the left, the laser on the right, and the low power IR remote on the top, which would contain the laser, in practice. (<b>b</b>) The cascoded header prototype in a 48 V power over ethernet PoE application.</p> Full article ">Figure 14
<p>The microwatt wake-up radio can share an antenna with the high-power transceiver.</p> Full article ">Figure 15
<p>Block diagram for how the WuR can be used to reduce standby power consumption in plug loads. The supply capacitor and burst mode logic provide a constant supply of microamp current to the wake-up radio. If the WuR needs to communicate with the main device, the circuit will require a header switch.</p> Full article ">Figure 16
<p>Prototype of the WuR method for standby power reduction. Includes the transmitter (right), receiver (bottom left), and display load (top). All of the receiver board components can be integrated into the WuR except for the electrolytic supply capacitor.</p> Full article ">
20 pages, 900 KiB  
Article
Co-Regulated Consensus of Cyber-Physical Resources in Multi-Agent Unmanned Aircraft Systems
by Chandima Fernando, Carrick Detweiler and Justin Bradley
Electronics 2019, 8(5), 569; https://doi.org/10.3390/electronics8050569 - 23 May 2019
Cited by 7 | Viewed by 3011
Abstract
Intelligent utilization of resources and improved mission performance in an autonomous agent require consideration of cyber and physical resources. The allocation of these resources becomes more complex when the system expands from one agent to multiple agents, and the control shifts from centralized [...] Read more.
Intelligent utilization of resources and improved mission performance in an autonomous agent require consideration of cyber and physical resources. The allocation of these resources becomes more complex when the system expands from one agent to multiple agents, and the control shifts from centralized to decentralized. Consensus is a distributed algorithm that lets multiple agents agree on a shared value, but typically does not leverage mobility. We propose a coupled consensus control strategy that co-regulates computation, communication frequency, and connectivity of the agents to achieve faster convergence times at lower communication rates and computational costs. In this strategy, agents move towards a common location to increase connectivity. Simultaneously, the communication frequency is increased when the shared state error between an agent and its connected neighbors is high. When the shared state converges (i.e., consensus is reached), the agents withdraw to the initial positions and the communication frequency is decreased. Convergence properties of our algorithm are demonstrated under the proposed co-regulated control algorithm. We evaluated the proposed approach through a new set of cyber-physical, multi-agent metrics and demonstrated our approach in a simulation of unmanned aircraft systems measuring temperatures at multiple sites. The results demonstrate that, compared with fixed-rate and event-triggered consensus algorithms, our co-regulation scheme can achieve improved performance with fewer resources, while maintaining high reactivity to changes in the environment and system. Full article
(This article belongs to the Special Issue Cyber-Physical Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Overview of the experimental design.</p> Full article ">Figure 2
<p>Behavior of the co-regulated algorithm.</p> Full article ">Figure 3
<p>Communication frequency with optimal alpha parameters, <math display="inline"><semantics> <mrow> <msubsup> <mi>α</mi> <mn>1</mn> <mi>F</mi> </msubsup> <mo>=</mo> <mn>1</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <msubsup> <mi>α</mi> <mn>2</mn> <mi>F</mi> </msubsup> <mo>=</mo> <mn>10</mn> </mrow> </semantics></math>.</p> Full article ">Figure 4
<p>Communication frequency at small alpha parameters, <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mn>1</mn> </msub> <mo>=</mo> <mn>0.05</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mn>2</mn> </msub> <mo>=</mo> <mn>0.05</mn> </mrow> </semantics></math>.</p> Full article ">
26 pages, 5712 KiB  
Article
Integrated Building Cells for a Simple Modular Design of Electronic Circuits with Reduced External Complexity: Performance, Active Element Assembly, and an Application Example
by Roman Sotner, Jan Jerabek, Ladislav Polak, Roman Prokop and Vilem Kledrowetz
Electronics 2019, 8(5), 568; https://doi.org/10.3390/electronics8050568 - 22 May 2019
Cited by 17 | Viewed by 4281
Abstract
This paper introduces new integrated analog cells fabricated in a C035 I3T25 0.35-μm ON Semiconductor process suitable for a modular design of advanced active elements with multiple terminals and controllable features. We developed and realized five analog cells on a single integrated circuit [...] Read more.
This paper introduces new integrated analog cells fabricated in a C035 I3T25 0.35-μm ON Semiconductor process suitable for a modular design of advanced active elements with multiple terminals and controllable features. We developed and realized five analog cells on a single integrated circuit (IC), namely a voltage differencing differential buffer, a voltage multiplier with current output in full complementary metal–oxide–semiconductor (CMOS) form, a voltage multiplier with current output with a bipolar core, a current-controlled current conveyor of the second generation with four current outputs, and a single-input and single-output adjustable current amplifier. These cells (sub-blocks of the manufactured IC device), designed to operate in a bandwidth of up to tens of MHz, can be used as a construction set for building a variety of advanced active elements, offering up to four independently adjustable internal parameters. The performances of all individual cells were verified by extensive laboratory measurements, and the obtained results were compared to simulations in the Cadence IC6 tool. The definition and assembly of a newly specified advanced active element, namely a current-controlled voltage differencing current conveyor transconductance amplifier (CC-VDCCTA), is shown as an example of modular interconnection of the selected cells. This device was implemented in a newly synthesized topology of an electronically linearly tunable quadrature oscillator. Features of this active element were verified by simulations and experimental measurements. Full article
(This article belongs to the Section Circuit and Signal Processing)
►▼ Show Figures

Figure 1

Figure 1
<p>The fabricated IC: (left) contents on a cell level and (right) an illustration of the top layout design.</p> Full article ">Figure 2
<p>Schematic symbol and interterminal transfer relation of the voltage differencing differential buffer (VDDB).</p> Full article ">Figure 3
<p>Schematic symbol and description of the ideal interterminal transfer relation of the voltage multiplier to the current output CMOS MLT (left) and BJT MLT (right).</p> Full article ">Figure 4
<p>Schematic symbol and description of the ideal interterminal transfer relations of the current-controlled current conveyor of the second generation (CCCII).</p> Full article ">Figure 5
<p>Schematic symbol and description of the ideal interterminal transfer relations of the adjustable current amplifier (CA).</p> Full article ">Figure 6
<p>Selected results of the measured and simulated responses of the VDDB: (left) DC transfer responses <span class="html-italic">Y</span>1–3 → <span class="html-italic">W</span>; (right) magnitude AC transfer responses <span class="html-italic">Y</span>1–3 → <span class="html-italic">W</span>.</p> Full article ">Figure 7
<p>Selected results of the measured and simulated responses of the CMOS MLT: (left) DC transfer responses <span class="html-italic">Y</span><sub>1</sub> → <span class="html-italic">Z</span> for a <span class="html-italic">V</span><sub>X1</sub> controlled by DC voltage; (right) magnitude of AC transfer responses <span class="html-italic">Y</span><sub>1</sub> → <span class="html-italic">Z</span> for a <span class="html-italic">V</span><sub>X1</sub> controlled by DC voltage.</p> Full article ">Figure 8
<p>Selected results of the measured and simulated responses of the BJT MLT: (left) DC transfer responses <span class="html-italic">X</span><sub>1</sub> → <span class="html-italic">Z</span> for a <span class="html-italic">V</span><sub>Y1</sub> controlled by DC voltage; (right) magnitude of AC transfer responses <span class="html-italic">X</span><sub>1</sub> → <span class="html-italic">Z</span> for a <span class="html-italic">V</span><sub>Y1</sub> controlled by DC voltage.</p> Full article ">Figure 9
<p>Selected results of the measured and simulated responses of the CCCII: (left) DC response of the <span class="html-italic">Y</span> → <span class="html-italic">X</span> transfer; (right) AC responses of <span class="html-italic">X</span> → <span class="html-italic">z</span><sub>1–2</sub> transfers.</p> Full article ">Figure 10
<p>Selected results of the measured and simulated responses of the CA: (left) AC responses of <span class="html-italic">i</span> → <span class="html-italic">o</span> transfers; (right) dependence of <span class="html-italic">B</span> on <span class="html-italic">I</span><sub>set_B</sub>.</p> Full article ">Figure 11
<p>Example of the interconnection of three cells of the fabricated IC, defining an advanced active element (AE) with three adjustable parameters: a so-called current-controlled voltage differencing current conveyor transconductance amplifier (CC-VDCCTA).</p> Full article ">Figure 12
<p>A simple electronically and linearly tunable quadrature oscillator based on CC-VDCCTA.</p> Full article ">Figure 13
<p>PCB for experimental verification of applications with fabricated chips shown in case of using only one IC package as discussed in the paper (implementation of the designed oscillator).</p> Full article ">Figure 14
<p>System for amplitude stabilization of the designed oscillator used in the experiments.</p> Full article ">Figure 15
<p>Experimental results for the oscillator: (left) output waveforms; (middle) spectral analysis of <span class="html-italic">V</span><sub>1</sub>; (right) spectral analysis of <span class="html-italic">V</span><sub>2</sub>.</p> Full article ">Figure 16
<p>Measured dependences with linear tuning of the frequency of oscillation (FO) of the oscillator: (left) <span class="html-italic">f</span><sub>0</sub> versus simultaneous control of <span class="html-italic">g</span><sub>m1</sub> (<span class="html-italic">V</span><sub>set_gm1</sub>) and <span class="html-italic">R</span><sub>x</sub> (<span class="html-italic">I</span><sub>set_Rx</sub>); (right) amplitude levels of <span class="html-italic">V</span><sub>1</sub> and <span class="html-italic">V</span><sub>2</sub> versus <span class="html-italic">f</span><sub>0</sub>.</p> Full article ">Figure A1
<p>Full CMOS topology of the voltage differencing differential buffer (VDDB).</p> Full article ">Figure A2
<p>Full CMOS topology of the voltage multiplier with current output (CMOS MLT).</p> Full article ">Figure A3
<p>Full CMOS topology of the voltage multiplier with current output (BJT MLT).</p> Full article ">Figure A4
<p>Full CMOS topology of the current-controlled current conveyor of the second generation (CCCII).</p> Full article ">Figure A5
<p>Full CMOS topology of the adjustable current amplifier (CA).</p> Full article ">
17 pages, 4363 KiB  
Article
Development of a Stand-Alone Photovoltaic System Considering Shaded Effect for Energy Storage and Release
by Kuei-Hsiang Chao, Yu-Ju Lai and Wen-Ching Chang
Electronics 2019, 8(5), 567; https://doi.org/10.3390/electronics8050567 - 22 May 2019
Cited by 4 | Viewed by 3088
Abstract
The purpose of this study was to develop a photovoltaic system that stores energy for use in direct current micro-grid systems or to supply electric power to consumers living in remote areas. If the photovoltaic module array is shaded, the signals of conventional [...] Read more.
The purpose of this study was to develop a photovoltaic system that stores energy for use in direct current micro-grid systems or to supply electric power to consumers living in remote areas. If the photovoltaic module array is shaded, the signals of conventional maximum power point trackers (MPPT) may be trapped at the local power maxima. Therefore, this study developed a smart maximum power point tracker to track the maximum power point (MPP). The control method adopted a teaching learning based optimization (TLBO) algorithm. To adjust the energy flow direction of the direct current load terminal, this study proposed an energy accumulation and release strategy that used a high-boost/buck-ratio bidirectional converter to control the battery charge and discharge for energy accumulation and release. In addition, this study developed an inverter to convert direct current into alternating current for alternating current loads. Full article
(This article belongs to the Special Issue Advances in Power Electronics Technologies for Renewable Energy Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>System architecture of the proposed photovoltaic system.</p> Full article ">Figure 2
<p>Architecture of the proposed maximum power point trackers (MPPT).</p> Full article ">Figure 3
<p>Architecture of the proposed high-boost/buck-ratio bidirectional converter.</p> Full article ">Figure 4
<p>Output power–voltage (P–V) characteristic curves of photovoltaic module array with different numbers of shaded panels.</p> Full article ">Figure 5
<p>Flow chart of the proposed teaching learning based optimization (TLBO) method applied to the photovoltaic module array for maximum power point tracking.</p> Full article ">Figure 6
<p>Photo of proposed circuit.</p> Full article ">Figure 7
<p>Output P–V curve of Case 1 photovoltaic module array.</p> Full article ">Figure 8
<p>Case 1 measured results of maximum power point tracking by using TLBO algorithm.</p> Full article ">Figure 9
<p>Output P–V curve of Case 2 photovoltaic module array.</p> Full article ">Figure 10
<p>Case 2 measured results of maximum power point tracking by using TLBO algorithm.</p> Full article ">Figure 11
<p>Output P–V curve of Case 3 photovoltaic module array.</p> Full article ">Figure 12
<p>Case 3 measured results of maximum power point tracking by using TLBO algorithm.</p> Full article ">Figure 13
<p>Output P–V curve of Case 4 photovoltaic module array.</p> Full article ">Figure 14
<p>Case 4 measured results of maximum power point tracking by using TLBO algorithm.</p> Full article ">Figure 15
<p>Measured results of the load power changed from 100 to 300 W and then lowered to 100 W.</p> Full article ">Figure 16
<p>Output P–V curves of the photovoltaic module array being unshaded and shaded.</p> Full article ">Figure 17
<p>Measured results of the battery under energy storage mode and auxiliary power supply mode.</p> Full article ">Figure 18
<p>Waveform of the alternating current (AC) power converted from the DC power.</p> Full article ">
12 pages, 4086 KiB  
Article
Underwater Robot Detection System Based on Fish’s Lateral Line
by Zhijie Tang, Zhen Wang, Jiaqi Lu, Gaoqian Ma and Pengfei Zhang
Electronics 2019, 8(5), 566; https://doi.org/10.3390/electronics8050566 - 22 May 2019
Cited by 16 | Viewed by 5184
Abstract
This paper introduces the near-field detection system of an underwater robot based on the fish lateral line. Inspired by the perception mechanism of fish’s lateral line, the aim is to add near-field detection functionality to an underwater vehicle. To mimic the fish’s lateral [...] Read more.
This paper introduces the near-field detection system of an underwater robot based on the fish lateral line. Inspired by the perception mechanism of fish’s lateral line, the aim is to add near-field detection functionality to an underwater vehicle. To mimic the fish’s lateral line, an array of pressure sensors is developed and installed on the surface of the underwater vehicle. A vibrating sphere is simulated as an underwater pressure source, and the moving mechanism is built to drive the sphere to vibrate at a certain frequency near the lateral line. The calculation of the near-field pressure generated by the vibrating sphere is derived by linearizing the kinematics and dynamics conditions of the free surface wave equation. Structurally, the geometry shape of the detection system is printed by a 3D printer. The pressure data are sent to the computer and analyzed immediately to obtain information of the pressure source. Through the experiment, the variation law of the pressure is generated when the source vibrates near the body, and is consistent with the simulation results of the derived pressure calculation formula. It is found that the direction of the near-field pressure source can distinguished. The pressure amplitude of the sampled signals are extracted to be prepared for the next step to estimate the vertical distance between the center of the pressure source and the lateral line. Full article
(This article belongs to the Special Issue Underwater Communication and Networking Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>The application scenes of underwater vehicles with lateral sensing systems.</p> Full article ">Figure 2
<p>Fish’s lateral line.</p> Full article ">Figure 3
<p>The fish-shaped prototype inspired by the trout lateral line. (<b>a</b>) Side view of the model; and (<b>b</b>) Top view of the model.</p> Full article ">Figure 4
<p>(<b>a</b>) The block diagram of the whole experimental principle. (<b>b</b>) The picture of the experimental setup.</p> Full article ">Figure 5
<p>Illustration of the problem formulation.</p> Full article ">Figure 6
<p>The control system block diagram.</p> Full article ">Figure 7
<p>Pressure sensor MS5803-05BA.</p> Full article ">Figure 8
<p>The software flow chart for reading pressure data.</p> Full article ">Figure 9
<p>Experimental diagram.</p> Full article ">Figure 10
<p>Pressure curve measured (<b>a</b>) and after spline fitting (<b>b</b>) of the eight sensors, respectively.</p> Full article ">Figure 11
<p>Pressure simulation curve of an individual node.</p> Full article ">Figure 12
<p>The measured pressure curve (<b>a</b>) and spline curve (<b>b</b>) of an individual sensor.</p> Full article ">Figure 13
<p>Simulation curve of pressure distribution of 8 nodes on the lateral line.</p> Full article ">Figure 14
<p>The measured pressure distribution curve (<b>a</b>) and spline curve (<b>b</b>) of lateral line sensor.</p> Full article ">Figure 15
<p>Frequency-amplitude diagram of seventh sensor.</p> Full article ">
14 pages, 4247 KiB  
Article
Gold/Polyimide-Based Resistive Strain Sensors
by Tao Han, Anindya Nag, Nasrin Afsarimanesh, Fowzia Akhter, Hangrui Liu, Samta Sapra, Subhas Mukhopadhyay and Yongzhao Xu
Electronics 2019, 8(5), 565; https://doi.org/10.3390/electronics8050565 - 22 May 2019
Cited by 36 | Viewed by 5288
Abstract
This paper presents the fabrication and implementation of novel resistive sensors that were implemented for strain-sensing applications. Some of the critical factors for the development of resistive sensors are addressed in this paper, such as the cost of fabrication, the steps of the [...] Read more.
This paper presents the fabrication and implementation of novel resistive sensors that were implemented for strain-sensing applications. Some of the critical factors for the development of resistive sensors are addressed in this paper, such as the cost of fabrication, the steps of the fabrication process which make it time-consuming to complete each prototype, and the inability to achieve optimised electrical and mechanical characteristics. The sensors were fabricated via magnetron sputtering of thin-film chromium and gold layer on the thin-film substrates at defined thicknesses. Sticky copper tapes were attached on the two sides of the sensor patches to form the electrodes. The operating principle of the fabricated sensors was based on the change in their responses with respect to the corresponding changes in their relative resistance as a function of the applied strain. The strain-induced characteristics of the patches were studied with different kinds of experiments, such as consecutive bending and pressure application. The sensors with 400 nm thickness of gold layer obtained a sensitivity of 0.0086 Ω/ppm for the pressure ranging between 0 and 400 kPa. The gauge factor of these sensors was between 4.9–6.6 for temperatures ranging between 25 °C and 55 °C. They were also used for tactile sensing to determine their potential as thin-film sensors for industrial applications, like in robotic and pressure-mapping applications. The results were promising in regards to the sensors’ controllable film thickness, easy operation, purity of the films and mechanically sound nature. These sensors can provide a podium to enhance the usage of resistive sensors on a higher scale to develop thin-film sensors for industrial applications. Full article
(This article belongs to the Special Issue Flexible/Stretchable Electronics)
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic diagram of the steps of fabrication for the gold-polyimide sensors. (<b>a</b>) After attaching the polyimide films onto the acrylic dishes with tapes, (<b>b</b>) the thin-films were sputtered with chromium and gold to form the sensing area of the samples. (<b>c</b>,<b>d</b>) The films were then taken off the template and attached to copper tapes on two sides to form the electrodes of the sensor patches.</p> Full article ">Figure 2
<p>The polyimide film with a thickness of 8 microns on an acrylic dish sputtered with chromium and gold.</p> Full article ">Figure 3
<p>SEM image of the sputtered sensor. The zoomed view shows the quality of the sputtering on the sensing surface.</p> Full article ">Figure 4
<p>Comparison of the size of the finished sample with an AUD 20 cents.</p> Full article ">Figure 5
<p>The working principle of the resistive sensors. The sensor consists of a variable resistor, which is the piezoresistive sensing element [<a href="#B42-electronics-08-00565" class="html-bibr">42</a>]. The resistive values of this element vary according to the magnitude and direction of the pressure applied to the sensors.</p> Full article ">Figure 6
<p>Consecutive bending of the gold-polyimide sensors to determine the repeatability in the changes in their responses. The inset of the figure shows the (<b>a</b>) normal and (<b>b</b>) bent positions.</p> Full article ">Figure 7
<p>The response of the sensors for the change in pressure ranging between 0 and 400 kPa. The experiments were done on three types of prototypes differing in the thickness of the gold layer for 200 nm, 300 nm and 400 nm.</p> Full article ">Figure 8
<p>The variation of relative resistance with respect to pressure for sensors with 400 nm. The sensitivity of the sensors for this range was 0.0086 Ω/kPa.</p> Full article ">Figure 9
<p>The response of the sensors for the change in temperature ranging between 30 °C and 85 °C.</p> Full article ">Figure 10
<p>The response of the sensors with respect to the change in pressure for four different temperatures of 25 °C, 35 °C and 45 °C and 55 °C.</p> Full article ">Figure 11
<p>The response of the sensors in terms of change in relative resistance for the applied strain.</p> Full article ">Figure 12
<p>Tactile sensing experiments performed on the developed patches. The inset of the figure shows the positioning of the index finger on the sensing area of the patches.</p> Full article ">Figure 13
<p>The response of the sensor patches with respect to the variation of tensile strain applied to them. The stretching of the sensors was varied from little stretched to full stretched condition to determine the increase in the corresponding resistance values.</p> Full article ">Figure 14
<p>Change in current with respect to different degrees of tensile strain.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-125
Previous Issue
Next Issue
Back to TopTop