Electronics

19 pages, 582 KiB

Open AccessArticle

Electrical Power Edge-End Interaction Modeling with Time Series Label Noise Learning

by Zhenshang Wang, Mi Zhou, Yuming Zhao, Fan Zhang, Jing Wang, Bin Qian, Zhen Liu, Peitian Ma and Qianli Ma

Electronics 2023, 12(18), 3987; https://doi.org/10.3390/electronics12183987 - 21 Sep 2023

Cited by 3 | Viewed by 1275

Abstract

In the context of electrical power systems, modeling the edge-end interaction involves understanding the dynamic relationship between different components and endpoints of the system. However, the time series of electrical power obtained by user terminals often suffer from low-quality issues such as missing [...] Read more.

In the context of electrical power systems, modeling the edge-end interaction involves understanding the dynamic relationship between different components and endpoints of the system. However, the time series of electrical power obtained by user terminals often suffer from low-quality issues such as missing values, numerical anomalies, and noisy labels. These issues can easily reduce the robustness of data mining results for edge-end interaction models. Therefore, this paper proposes a time–frequency noisy label classification (TF-NLC) model, which improves the robustness of edge-end interaction models in dealing with low-quality issues. Specifically, we employ two deep neural networks that are trained concurrently, utilizing both the time and frequency domains. The two networks mutually guide each other’s classification training by selecting clean labels from batches within small loss data. To further improve the robustness of the classification of time and frequency domain feature representations, we introduce a time–frequency domain consistency contrastive learning module. By classifying the selection of clean labels based on time–frequency representations for mutually guided training, TF-NLC can effectively mitigate the negative impact of noisy labels on model training. Extensive experiments on eight electrical power and ten other different realistic scenario time series datasets show that our proposed TF-NLC achieves advanced classification performance under different noisy label scenarios. Also, the ablation and visualization experiments further demonstrate the robustness of our proposed method. Full article

(This article belongs to the Special Issue Knowledge Engineering and Data Mining Volume II)

► Show Figures

Figure 1

Figure 1
The framework for the time–frequency noisy label classification model. Full article ">Figure 2
The left plot shows the bluecurve representing a time series in C1P1 that is labeled as normal electricity consumption behavior. On the right plot, the red curve represents a series in C1P1 that is labeled as abnormal electricity consumption behavior (notably around 9 o’clock, where there is an abnormal fluctuation in the sensor readings). Full article ">Figure 3
The t-SNE visualization depicts the performance of three methods on the Epilepsy dataset under 30% symmetric noise. Four classes are shown in different colors. Among these visualizations: (a) illustrates the data feature representation learned by one of the encoders after training, where the dual network (TT) works in the time domain; (b) shows the data feature representation obtained by one of the encoders after training, with the dual network (FF) working in the frequency domain; (c) demonstrates the data feature representation learned by the encoder (T) working in the time domain for TF-NLC; (d) indicates the data feature representation learned by the encoder (F) working in the frequency domain for TF-NLC. Upon comparing (a) with (c) and (b) with (d), it can be found that the feature representation learned by TF-NLC in both the time and frequency domains exhibits greater robustness as the feature representation is more compact. Full article ">Figure 4
The training loss curves of the two methods on the ArrowHead dataset under the 30% symmetric noise setting are displayed. Specifically, (a) represents the loss curve for Vanilla, and (b) illustrates the loss curve for TF-NLC. The red curve corresponds to the average training loss of the noisy samples, while the blue curve represents the average training loss of the clean samples. Full article ">

23 pages, 807 KiB

Open AccessFeature PaperArticle

Open-Source HW/SW Co-Simulation Using QEMU and GHDL for VHDL-Based SoC Design

by Giorgio Biagetti, Laura Falaschetti, Paolo Crippa, Michele Alessandrini and Claudio Turchetti

Electronics 2023, 12(18), 3986; https://doi.org/10.3390/electronics12183986 - 21 Sep 2023

Cited by 3 | Viewed by 4348

Abstract

Hardware/software co-simulation is a technique that can help design and validate digital circuits controlled by embedded processors. Co-simulation has largely been applied to system-level models, and tools for SystemC or SystemVerilog are readily available, but they are either not compatible or very cumbersome [...] Read more.

Hardware/software co-simulation is a technique that can help design and validate digital circuits controlled by embedded processors. Co-simulation has largely been applied to system-level models, and tools for SystemC or SystemVerilog are readily available, but they are either not compatible or very cumbersome to use with VHDL, the most commonly used language for FPGA design. This paper presents a direct, simple-to-use solution to co-simulate a VHDL design together with the firmware (FW) that controls it. It aims to bring the power of co-simulation to every digital designer, so it uses open-source tools, and the developed code is also open. A small patch applied to the QEMU emulator allows it to communicate with a custom-written VHDL module that exposes a CPU bus to the digital design, controlled by the FW emulated in QEMU. No changes to FW code or VHDL device code are required: with our approach, it is possible to co-simulate the very same code base that would then be implemented into an FPGA, enabling debugging, verification, and tracing capabilities that would not be possible even with the real hardware. Full article

(This article belongs to the Special Issue Embedded Systems: Fundamentals, Design and Practical Applications)

► Show Figures

Figure 1

21 pages, 4868 KiB

Open AccessArticle

Optimizing Long Short-Term Memory Network for Air Pollution Prediction Using a Novel Binary Chimp Optimization Algorithm

by Sahba Baniasadi, Reza Salehi, Sepehr Soltani, Diego Martín, Parmida Pourmand and Ehsan Ghafourian

Electronics 2023, 12(18), 3985; https://doi.org/10.3390/electronics12183985 - 21 Sep 2023

Cited by 5 | Viewed by 2000

Abstract

Elevated levels of fine particulate matter (PM_2.5) in the atmosphere present substantial risks to human health and welfare. The accurate assessment of PM_2.5 concentrations plays a pivotal role in facilitating prompt responses by pertinent regulatory bodies to mitigate air pollution. [...] Read more.

Elevated levels of fine particulate matter (PM_2.5) in the atmosphere present substantial risks to human health and welfare. The accurate assessment of PM_2.5 concentrations plays a pivotal role in facilitating prompt responses by pertinent regulatory bodies to mitigate air pollution. Additionally, it furnishes indispensable information for epidemiological studies concentrating on PM_2.5 exposure. In recent years, predictive models based on deep learning (DL) have offered promise in improving the accuracy and efficiency of air quality forecasts when compared to other approaches. Long short-term memory (LSTM) networks have proven to be effective in time series forecasting tasks, including air pollution prediction. However, optimizing LSTM models for enhanced accuracy and efficiency remains an ongoing research area. In this paper, we propose a novel approach that integrates the novel binary chimp optimization algorithm (BChOA) with LSTM networks to optimize air pollution prediction models. The proposed BChOA, inspired by the social behavior of chimpanzees, provides a powerful optimization technique to fine-tune the LSTM architecture and optimize its parameters. The evaluation of the results is performed using cross-validation methods such as the coefficient of determination (

R^{2}

), accuracy, the root mean square error (RMSE), and receiver operating characteristic (ROC) curve. Additionally, the performance of the BChOA-LSTM model is compared against eight DL architectures. Experimental evaluations using real-world air pollution data demonstrate the superior performance of the proposed BChOA-based LSTM model compared to traditional LSTM models and other optimization algorithms. The BChOA-LSTM model achieved the highest accuracy of 96.41% on the validation datasets, making it the most successful approach. The results show that the BChOA-LSTM architecture performs better than the other architectures in terms of the

R^{2}

convergence curve, RMSE, and accuracy. Full article

(This article belongs to the Special Issue Advances in Embedded Deep Learning Systems)

► Show Figures

Figure 1

14 pages, 536 KiB

Open AccessArticle

Towards Privacy-Preserving Federated Neuromorphic Learning via Spiking Neuron Models

by Bing Han, Qiang Fu and Xinliang Zhang

Electronics 2023, 12(18), 3984; https://doi.org/10.3390/electronics12183984 - 21 Sep 2023

Cited by 2 | Viewed by 1628

Abstract

Federated learning (FL) has been broadly adopted in both academia and industry in recent years. As a bridge to connect the so-called “data islands”, FL has contributed greatly to promoting data utilization. In particular, FL enables disjoint entities to cooperatively train a shared [...] Read more.

Federated learning (FL) has been broadly adopted in both academia and industry in recent years. As a bridge to connect the so-called “data islands”, FL has contributed greatly to promoting data utilization. In particular, FL enables disjoint entities to cooperatively train a shared model, while protecting each participant’s data privacy. However, current FL frameworks cannot offer privacy protection and reduce the computation overhead at the same time. Therefore, its implementation in practical scenarios, such as edge computing, is limited. In this paper, we propose a novel FL framework with spiking neuron models and differential privacy, which simultaneously provides theoretically guaranteed privacy protection and achieves low energy consumption. We model the local forward propagation process in a discrete way similar to nerve signal travel in the human brain. Since neurons only fire when the accumulated membrane potential exceeds a threshold, spiking neuron models require significantly lower energy compared to traditional neural networks. In addition, to protect sensitive information in model gradients, we add differently private noise in both the local training phase and server aggregation phase. Empirical evaluation results show that our proposal can effectively reduce the accuracy of membership inference attacks and property inference attacks, while maintaining a relatively low energy cost. blueFor example, the attack accuracy of a membership inference attack drops to 43% in some scenarios. As a result, our proposed FL framework can work well in large-scale cross-device learning scenarios. Full article

(This article belongs to the Special Issue Data Privacy and Cybersecurity in Mobile Crowdsensing)

► Show Figures

Figure 1

12 pages, 16406 KiB

Open AccessArticle

A Study on Webtoon Generation Using CLIP and Diffusion Models

by Kyungho Yu, Hyoungju Kim, Jeongin Kim, Chanjun Chun and Pankoo Kim

Electronics 2023, 12(18), 3983; https://doi.org/10.3390/electronics12183983 - 21 Sep 2023

Viewed by 2769

Abstract

This study focuses on harnessing deep-learning-based text-to-image transformation techniques to help webtoon creators’ creative outputs. We converted publicly available datasets (e.g., MSCOCO) into a multimodal webtoon dataset using CartoonGAN. First, the dataset was leveraged for training contrastive language image pre-training (CLIP), a model [...] Read more.

This study focuses on harnessing deep-learning-based text-to-image transformation techniques to help webtoon creators’ creative outputs. We converted publicly available datasets (e.g., MSCOCO) into a multimodal webtoon dataset using CartoonGAN. First, the dataset was leveraged for training contrastive language image pre-training (CLIP), a model composed of multi-lingual BERT and a Vision Transformer that learnt to associate text with images. Second, a pre-trained diffusion model was employed to generate webtoons through text and text-similar image input. The webtoon dataset comprised treatments (i.e., textual descriptions) paired with their corresponding webtoon illustrations. CLIP (operating through contrastive learning) extracted features from different data modalities and aligned similar data more closely within the same feature space while pushing dissimilar data apart. This model learnt the relationships between various modalities in multimodal data. To generate webtoons using the diffusion model, the process involved providing the CLIP features of the desired webtoon’s text with those of the most text-similar image to a pre-trained diffusion model. Experiments were conducted using both single- and continuous-text inputs to generate webtoons. In the experiments, both single-text and continuous-text inputs were used to generate webtoons, and the results showed an inception score of 7.14 when using continuous-text inputs. The text-to-image technology developed here could streamline the webtoon creation process for artists by enabling the efficient generation of webtoons based on the provided text. However, the current inability to generate webtoons from multiple sentences or images while maintaining a consistent artistic style was noted. Therefore, further research is imperative to develop a text-to-image model capable of handling multi-sentence and -lingual input while ensuring coherence in the artistic style across the generated webtoon images. Full article

(This article belongs to the Special Issue Machine Learning and Deep Learning Based Pattern Recognition)

► Show Figures

Figure 1

16 pages, 3630 KiB

Open AccessTechnical Note

A Novel DME-YOLO Structure in a High-Frequency Transformer Improves the Accuracy and Speed of Detection

by Zhiqiang Kang, Wenqian Jiang, Lile He and Chenrui Zhang

Electronics 2023, 12(18), 3982; https://doi.org/10.3390/electronics12183982 - 21 Sep 2023

Cited by 1 | Viewed by 1801

Abstract

Traditional YOLO models face a dilemma when it comes to dim detection targets: the detection accuracy increases while the speed inevitably reduces, or vice versa. To resolve this issue, we propose a novel DME-YOLO model, which is characterized by the establishment of a [...] Read more.

Traditional YOLO models face a dilemma when it comes to dim detection targets: the detection accuracy increases while the speed inevitably reduces, or vice versa. To resolve this issue, we propose a novel DME-YOLO model, which is characterized by the establishment of a backbone based on the YOLOv7 and Dense blocks. Moreover, through the application of feature multiplexing, both the parameters and floating-point computation were decreased; therefore, the defect detection process was accelerated. We also designed a multi-source attention mechanism module called MSAM, which is capable of integrating spatial information from multiple sources. Due to its outstanding quality, the addition of MSAM as the neck of the original YOLOv7 model compensated for the loss of spatial information in the process of forward propagation, thereby improving the detection accuracy of small target defects and simultaneously ensuring real-time detection. Finally, EIOU was adopted as a loss function to bolster the target frame regression process. The results of the experiment indicated detection accuracy and speed values of up to 97.6 mAP and 51.2 FPS, respectively, suggesting the superiority of the model. Compared with the YOLOv7 model, the experimental parameters for the novel DME-YOLO increased by 2.8% for mAP and 15.7 for FPS, respectively. In conclusion, the novel DME-YOLO model had excellent overall performance regarding detection speed and accuracy. Full article

(This article belongs to the Special Issue Advances and Applications of Computer Vision in Electronics)

► Show Figures

Figure 1

17 pages, 4959 KiB

Open AccessArticle

Target Localization and Grasping of NAO Robot Based on YOLOv8 Network and Monocular Ranging

by Yingrui Jin, Zhaoyuan Shi, Xinlong Xu, Guang Wu, Hengyi Li and Shengjun Wen

Electronics 2023, 12(18), 3981; https://doi.org/10.3390/electronics12183981 - 21 Sep 2023

Cited by 10 | Viewed by 2522

Abstract

As a typical visual positioning system, monocular ranging is widely used in various fields. However, when the distance increases, there is a greater error. YOLOv8 network has the advantages of fast recognition speed and high accuracy. This paper proposes a method by combining [...] Read more.

As a typical visual positioning system, monocular ranging is widely used in various fields. However, when the distance increases, there is a greater error. YOLOv8 network has the advantages of fast recognition speed and high accuracy. This paper proposes a method by combining YOLOv8 network recognition with a monocular ranging method to achieve target localization and grasping for the NAO robots. By establishing a visual distance error compensation model and applying it to correct the estimation results of the monocular distance measurement model, the accuracy of the NAO robot’s long-distance monocular visual positioning is improved. Additionally, a grasping control strategy based on pose interpolation is proposed. Throughout, the proposed method’s advantage in measurement accuracy was confirmed via experiments, and the grasping strategy has been implemented to accurately grasp the target object. Full article

► Show Figures

Figure 1

19 pages, 3408 KiB

Open AccessArticle

Convolutional Neural Networks Adapted for Regression Tasks: Predicting the Orientation of Straight Arrows on Marked Road Pavement Using Deep Learning and Rectified Orthophotography

by Calimanut-Ionut Cira, Alberto Díaz-Álvarez, Francisco Serradilla and Miguel-Ángel Manso-Callejo

Electronics 2023, 12(18), 3980; https://doi.org/10.3390/electronics12183980 - 21 Sep 2023

Cited by 5 | Viewed by 3896

Abstract

Arrow signs found on roadway pavement are an important component of modern transportation systems. Given the rise in autonomous vehicles, public agencies are increasingly interested in accurately identifying and analysing detailed road pavement information to generate comprehensive road maps and decision support systems [...] Read more.

Arrow signs found on roadway pavement are an important component of modern transportation systems. Given the rise in autonomous vehicles, public agencies are increasingly interested in accurately identifying and analysing detailed road pavement information to generate comprehensive road maps and decision support systems that can optimise traffic flow, enhance road safety, and provide complete official road cartographic support (that can be used in autonomous driving tasks). As arrow signs are a fundamental component of traffic guidance, this paper aims to present a novel deep learning-based approach to identify the orientation and direction of arrow signs on marked roadway pavements using high-resolution aerial orthoimages. The approach is based on convolutional neural network architectures (VGGNet, ResNet, Xception, and DenseNet) that are modified and adapted for regression tasks with a proposed learning structure, together with an ad hoc model, specially introduced for this task. Although the best-performing artificial neural network was based on VGGNet (VGG-19 variant), it only slightly surpassed the proposed ad hoc model in the average values of the R² score, mean squared error, and angular error by 0.005, 0.001, and 0.036, respectively, using the training set (the ad hoc model delivered an average R² score, mean squared error, and angular error of 0.9874, 0.001, and 2.516, respectively). Furthermore, the ad hoc model’s predictions using the test set were the most consistent (a standard deviation of the R² score of 0.033 compared with the score of 0.042 achieved using VGG19), while being almost eight times more computationally efficient when compared with the VGG19 model (2,673,729 parameters vs VGG19′s 20,321,985 parameters). Full article

(This article belongs to the Special Issue Advances in Computer Vision and Deep Learning and Its Applications)

► Show Figures

Figure 1

21 pages, 6773 KiB

Open AccessFeature PaperArticle

Comparison of 2L + 2M and 6L SVPWM for Five-Phase Inverter to Reduce Common Mode Voltage

by Kotb B. Tawfiq, Arafa S. Mansour and Peter Sergeant

Electronics 2023, 12(18), 3979; https://doi.org/10.3390/electronics12183979 - 21 Sep 2023

Cited by 2 | Viewed by 1516

Abstract

Multiphase drives have received a lot of interest because of their several features over traditional three-phase systems for high-power applications. Pulse-width modulation (PWM) approaches are necessary to regulate the supply for multiphase ac drives. As a result, it is vital to continually improve [...] Read more.

Multiphase drives have received a lot of interest because of their several features over traditional three-phase systems for high-power applications. Pulse-width modulation (PWM) approaches are necessary to regulate the supply for multiphase ac drives. As a result, it is vital to continually improve the modulation and control approaches used to upgrade output power converters’ quality. This paper offers a comparative analysis of the 2L + 2M and 6L space vector pulse-width modulation (SVPWM) techniques applied to a five-phase two-level voltage source inverter (VSI) fed an inductive (R-L) load. The comparative evaluation is based on measuring the inverter switching losses, the total harmonic distortion (THD) values, and the common mode voltage (CMV) under different operation scenarios. A system model is carried out by MATLAB/Simulink. An experimental prototype is constructed in the lab to validate the theoretical analysis. Simulation results for the system based on the two SVPWM techniques are obtained at different modulation indices and different output frequencies and are confirmed by the experimental results. It has been found that the peak-to-peak CMV of the 6L method is 80% lower than that of the 2L + 2M method. Moreover, 6L SVPWM offers better DC-link utilization compared to 2L + 2M SVPWM. Full article

(This article belongs to the Special Issue Power Electronic Converters in a Multiphase Drive Systems)

► Show Figures

Figure 1

19 pages, 9083 KiB

Open AccessArticle

A Deep Learning-Enhanced Stereo Matching Method and Its Application to Bin Picking Problems Involving Tiny Cubic Workpieces

by Masaru Yoshizawa, Kazuhiro Motegi and Yoichi Shiraishi

Electronics 2023, 12(18), 3978; https://doi.org/10.3390/electronics12183978 - 21 Sep 2023

Viewed by 1457

Abstract

This paper proposes a stereo matching method enhanced by object detection and instance segmentation results obtained through the use of a deep convolutional neural network. Then, this method is applied to generate a picking plan to solve bin picking problems, that is, to [...] Read more.

This paper proposes a stereo matching method enhanced by object detection and instance segmentation results obtained through the use of a deep convolutional neural network. Then, this method is applied to generate a picking plan to solve bin picking problems, that is, to automatically pick up objects with random poses in a stack using a robotic arm. The system configuration and bin picking process flow are suggested using the proposed method, and it is applied to bin picking problems, especially those involving tiny cubic workpieces. The picking plan is generated by applying the Harris corner detection algorithm to the point cloud in the generated three-dimensional map. In the experiments, two kinds of stacks consisting of cubic workpieces with an edge length of 10 mm or 5 mm are tested for bin picking. In the first bin picking problem, all workpieces are successfully picked up, whereas in the second, the depths of the workpieces are obtained, but the instance segmentation process is not completed. In future work, not only cubic workpieces but also other arbitrarily shaped workpieces should be recognized in various types of bin picking problems. Full article

(This article belongs to the Special Issue Recent Advances in Object Detection and Image Processing)

► Show Figures

Figure 1

14 pages, 1476 KiB

Open AccessArticle

Property Analysis of Gateway Refinement of Object-Oriented Petri Net with Inhibitor-Arcs-Based Representation for Embedded Systems

by Chuanliang Xia, Mengying Qin, Yan Sun and Maibo Guo

Electronics 2023, 12(18), 3977; https://doi.org/10.3390/electronics12183977 - 21 Sep 2023

Viewed by 1147

Abstract

This paper focuses on embedded system modeling, proposing a solution to obtain a refined net via the refinement operation of an extended Petri net. Object-oriented technology and Petri net with inhibitor-arcs-based representation for embedded systems (PIRES+) are combined to obtain an object-oriented PIRES+ [...] Read more.

This paper focuses on embedded system modeling, proposing a solution to obtain a refined net via the refinement operation of an extended Petri net. Object-oriented technology and Petri net with inhibitor-arcs-based representation for embedded systems (PIRES+) are combined to obtain an object-oriented PIRES+ (OOPIRES+). A gateway refinement method of OOPIRES+ is proposed, and the preservation of the liveness, boundedness, reachability, functionality, and timing of the refined net system is investigated. The modeling analysis of a smart home system is taken as an example to verify the effectiveness of the refinement method. The results can provide an effective way for the investigation of the refined properties of a Petri net system and a favorable means for large-scale complex embedded system modeling, which has broad application prospects. Full article

(This article belongs to the Special Issue Deep Learning for Data Mining: Theory, Methods, and Applications)

► Show Figures

Figure 1

17 pages, 5460 KiB

Open AccessArticle

An Objective Holographic Feedback Linearization Based on a Sliding Mode Control for a Buck Converter with a Constant Power Load

by Jiyong Li, Benquan Pi, Pengcheng Zhou, Jingwen Li, Hao Dong and Peiwen Chen

Electronics 2023, 12(18), 3976; https://doi.org/10.3390/electronics12183976 - 21 Sep 2023

Viewed by 1185

Abstract

As a typical load, the constant power load (CPL) has negative impedance characteristics. The stability of the buck converter system with a mixed load of CPL and resistive load is affected by the size of the CPL. When the resistive load is larger [...] Read more.

As a typical load, the constant power load (CPL) has negative impedance characteristics. The stability of the buck converter system with a mixed load of CPL and resistive load is affected by the size of the CPL. When the resistive load is larger than the CPL, the buck converter with the output voltage as an output function is a non-minimum phase nonlinear system, because its linear approximation has a right-half-plane pole. The non-minimum phase characteristic limits the application of many control techniques, but the objective holographic feedback linearization control (OHFLC) method is a good control strategy that can bypass the non-minimum phase system and make the system stable. However, the traditional OHFLC method, in designing the controller, generally uses a linear optimal quadratic design method to obtain a linear feedback control law. It requires a state quantity component with a one-order relative degree to the system. But it is not easy to find such a suitable state quantity with a one-order relative degree to the system. In this paper, an improved OHFLC method is proposed for Buck converters with a mixed loads of CPL and resistive loads, using the sliding mode control (SMC) theory to design the controller, so that the output state quantity components with different relative degrees to the system can be used in the holographic feedback linearization method. Finally, the simulation and experimental results also demonstrate that this method has the same, or even better, dynamic response performance and robustness than the traditional OHFLC method. Full article

(This article belongs to the Special Issue Advanced Control Techniques of Power Electronics)

► Show Figures

Figure 1

20 pages, 11052 KiB

Open AccessArticle

A High-Power Density DC Converter for Medium-Voltage DC Distribution Networks

by Dai Wan, Qianfan Zhou, Xujin Duan, Jiran Zhu, Junhao Li and Hengyi Zhou

Electronics 2023, 12(18), 3975; https://doi.org/10.3390/electronics12183975 - 21 Sep 2023

Cited by 3 | Viewed by 1719

Abstract

A DC converter is the core equipment of voltage conversion and power distribution in a DC distribution network. Its operating characteristics have a profound impact on the flexible regulation of distributed resources in an active distribution network. It is challenging for the existing [...] Read more.

A DC converter is the core equipment of voltage conversion and power distribution in a DC distribution network. Its operating characteristics have a profound impact on the flexible regulation of distributed resources in an active distribution network. It is challenging for the existing single-stage conversion topology to meet the requirements of distributed renewable energy connected to a multi-voltage level, medium-voltage grid. It is necessary to study the multistage transform power unit topology further, which can satisfy high reliability, high efficiency, and wide input range. This paper proposes a high-power density DC converter for medium-voltage DC networks with wide voltage levels. It adopts Buck-LLC integrated modular composition. The input ends of the high isolation resonant power unit are connected in series to provide high voltage endurance, and the output ends are connected in parallel to meet the high-power demand and achieve high-power transmission efficiency. The proposed series dual Buck-LLC resonant power unit topology can adjust the duty cycle of series dual buck circuits to meet the needs of different levels of medium-voltage DC power grids. The soft switching problem within the wide input range of all switching tubes is solved by introducing auxiliary inductors, thereby improving energy transmission efficiency. The auxiliary circuit and control parameters are optimized based on the research of each switching tube’s soft switching boundary conditions. Finally, an experimental prototype of a 6.25~7 kW power unit is designed and developed to prove the proposed topology’s feasibility and effectiveness. Great breakthroughs have been made both in theoretical research and engineering prototype development. Full article

(This article belongs to the Special Issue The Hybrid AC-DC Power System Coordinated Control and Operation Technology)

► Show Figures

Figure 1

15 pages, 1483 KiB

Open AccessArticle

Miniaturized Dual-Band SIW-Based Bandpass Filters Using Open-Loop Ring Resonators

by Nrusingha Charan Pradhan, Slawomir Koziel, Rusan Kumar Barik, Anna Pietrenko-Dabrowska and Sholampettai Subramanian Karthikeyan

Electronics 2023, 12(18), 3974; https://doi.org/10.3390/electronics12183974 - 21 Sep 2023

Cited by 12 | Viewed by 2400

Abstract

This article presents two novel architectures of dual-band substrate-integrated waveguide (SIW) bandpass filters (BPFs). Initially, two identical open-loop ring resonators (OLRRs) were coupled face-to-face on the top of the SIW cavity to realize a dual-band single-pole BPF. To obtain two-pole dual-band characteristics, two [...] Read more.

This article presents two novel architectures of dual-band substrate-integrated waveguide (SIW) bandpass filters (BPFs). Initially, two identical open-loop ring resonators (OLRRs) were coupled face-to-face on the top of the SIW cavity to realize a dual-band single-pole BPF. To obtain two-pole dual-band characteristics, two OLRRs resonant units were assembled horizontally within the top metal layer of the SIW, which is a technique used for the first time in the literature. For demonstration purposes, two types of SIW filters loaded with OLRRs were designed and fabricated. The proposed filters feature an extremely compact size, a low insertion loss, and good selectivity. The single- and two-pole filters have an overall size of 0.012

λ_{g}

^{2}

and 0.041

λ_{g}

^{2}

, respectively. The simulated and measured circuit responses are in good agreement. Full article

(This article belongs to the Special Issue Advanced RF, Microwave, and Millimeter-Wave Circuits and Systems)

► Show Figures

Figure 1

9 pages, 996 KiB

Open AccessArticle

Shielding Effectiveness of Unmanned Aerial Vehicle Electronics with Graphene-Based Absorber

by Roman Kubacki, Rafał Przesmycki and Dariusz Laskowski

Electronics 2023, 12(18), 3973; https://doi.org/10.3390/electronics12183973 - 21 Sep 2023

Cited by 7 | Viewed by 1727

Abstract

Within this study, we explored the augmented security measures for the electronics of unmanned aerial vehicles (UAVs) within an RF environment. UAVs are commonly utilised across various sectors, and their use as auxiliary platforms for cellular networks, as parallel networks working in tandem [...] Read more.

Within this study, we explored the augmented security measures for the electronics of unmanned aerial vehicles (UAVs) within an RF environment. UAVs are commonly utilised across various sectors, and their use as auxiliary platforms for cellular networks, as parallel networks working in tandem with ground-based base stations, holds considerable promise. In this context, ensuring the uninterrupted operation of UAVs is a paramount objective. However, the considerable external electromagnetic interference emitted by existing base stations may jeopardise the functionality of UAV electronics. This could potentially lead to an unintended flight path and a sudden cessation of communication with the operator. To mitigate the detrimental impact of the RF field, we advocate covering the UAV casing with reduced graphene oxide (RGO). The efficacy of RGO’s shielding effectiveness (SE) was investigated over a frequency spectrum from 100 MHz to 10 GHz. Our scrutiny of this property was centred around the measurement of scattering matrix coefficients of the unadulterated material—without additives of any kind. Our findings show that this material is a favourable candidate for UAV absorbers due to its low reflection coefficient coupled with its high absorption capacity. The studied absorber ensures an SE value of 25 dB and 30 dB for a 3 mm layer at frequencies of 3.6 GHz (pertaining to the 5G system) and 5.8 GHz (pertaining to LTE), respectively. Full article

(This article belongs to the Section Microwave and Wireless Communications)

► Show Figures

Figure 1

22 pages, 5465 KiB

Open AccessArticle

Model Predictive Secondary Frequency Control for Islanded Microgrid under Wind and Solar Stochastics

by Zhongwei Zhao, Xiangyu Zhang and Cheng Zhong

Electronics 2023, 12(18), 3972; https://doi.org/10.3390/electronics12183972 - 21 Sep 2023

Cited by 2 | Viewed by 1585

Abstract

As microgrids are the main carriers of renewable energy sources (RESs), research on them has been receiving more attention. When considering the increase in the penetration of renewable energy sources/distributed generators (DGs) in microgrids, their low inertia and high stochastic power disturbance pose [...] Read more.

As microgrids are the main carriers of renewable energy sources (RESs), research on them has been receiving more attention. When considering the increase in the penetration of renewable energy sources/distributed generators (DGs) in microgrids, their low inertia and high stochastic power disturbance pose more challenges for frequency control. To address these challenges, this paper proposes a model predictive control (MPC) secondary control that incorporates an unknown input observer and where RESs/DGs use a deloading virtual synchronous generator (VSG) control to improve the system’s inertia. An unknown input observer is employed to estimate the system states and random power disturbance from the RESs/DGs and load to improve the effect of the predictive control. The distributed restorative power of each DG is obtained by solving the quadratic programming (QP) optimal problem with variable constraints. The RESs/DGs are given priority to participate in secondary frequency control due to the proper weighting factors being set. An islanded microgrid model consisting of multiple photovoltaic and wind power sources was built. The simulation results demonstrate that the proposed method improves the system frequency, restoration speed, and reduces frequency deviations compared with the traditional secondary control method. Full article

► Show Figures

Figure 1

22 pages, 8506 KiB

Open AccessArticle

Anomaly Detection Methods for Industrial Applications: A Comparative Study

by Maria Antonietta Panza, Marco Pota and Massimo Esposito

Electronics 2023, 12(18), 3971; https://doi.org/10.3390/electronics12183971 - 20 Sep 2023

Cited by 8 | Viewed by 4122

Abstract

Anomaly detection (AD) algorithms can be instrumental in industrial scenarios to enhance the detection of potentially serious problems at a very early stage. Of course, the “Industry 4.0” revolution is fostering the implementation of intelligent data-driven decisions in industry based on increasingly efficient [...] Read more.

Anomaly detection (AD) algorithms can be instrumental in industrial scenarios to enhance the detection of potentially serious problems at a very early stage. Of course, the “Industry 4.0” revolution is fostering the implementation of intelligent data-driven decisions in industry based on increasingly efficient machine learning (ML) algorithms. Most well-known AD methods use a supervised learning approach focusing on fault classification. They assume the availability of labeled data for both normal and anomalous classes. However, in many industrial environments, a labeled set of anomalous data instances is more challenging to obtain than a labeled set of normal data. Hence, this work implements an unsupervised approach based on two different methods using a typical benchmark bearing-fault dataset. The first method relies on the manual extraction of typical vibration metrics provided as input to an ML algorithm. The second one is based on a deep learning (DL) approach, automatically learning latent representation from raw data. The performance metrics demonstrate that both approaches can distinguish the state of a bearing from normal to faulty. DL methodology proves a higher accuracy rate in recognizing faults and a better ability to provide information about the fault size. Full article

(This article belongs to the Special Issue Advances in Predictive Maintenance for Critical Infrastructure)

► Show Figures

Figure 1

11 pages, 436 KiB

Open AccessArticle

Python Framework for Modular and Parametric SPICE Netlists Generation

by Sergio Vinagrero Gutiérrez, Giorgio Di Natale and Elena-Ioana Vatajelu

Electronics 2023, 12(18), 3970; https://doi.org/10.3390/electronics12183970 - 20 Sep 2023

Cited by 1 | Viewed by 1956

Abstract

Due to the complex specifications of current electronic systems, design decisions need to be explored automatically. However, the exploration process is a complex task given the plethora of design choices such as the selection of components, number of components, operating modes of each [...] Read more.

Due to the complex specifications of current electronic systems, design decisions need to be explored automatically. However, the exploration process is a complex task given the plethora of design choices such as the selection of components, number of components, operating modes of each of the components, connections between the components and variety of ways in which the same functionality can be implemented. To tackle these issues, scripts are used to generate designs based on high-level abstract constructions. Still, this approach is usually ad hoc and platform dependent, making the whole procedure hardly reusable, scalable and versatile. We propose a generic, open-source framework tackling rapid design exploration for the generation of modular and parametric electronic designs that is able to work on any major simulator. Full article

(This article belongs to the Special Issue Design of Mixed Analog/Digital Circuits, Volume 2)

► Show Figures

Figure 1

15 pages, 11341 KiB

Open AccessArticle

YOLO v7-ECA-PConv-NWD Detects Defective Insulators on Transmission Lines

by Jianrui Zhang, Xia Wei, Linxuan Zhang, Libin Yu, Yannan Chen and Meiqi Tu

Electronics 2023, 12(18), 3969; https://doi.org/10.3390/electronics12183969 - 20 Sep 2023

Cited by 14 | Viewed by 2960

Abstract

This paper proposes an enhanced YOLO v7-based method for detecting insulator defects in transmission lines, addressing the challenges of low accuracy and high leakage rates caused by complex backgrounds and electric poles alongside varying sizes of insulator targets in the image. Firstly, to [...] Read more.

This paper proposes an enhanced YOLO v7-based method for detecting insulator defects in transmission lines, addressing the challenges of low accuracy and high leakage rates caused by complex backgrounds and electric poles alongside varying sizes of insulator targets in the image. Firstly, to address the issue of background interference and improve the importance of insulator features, a lightweight attention mechanism named Efficient Channel Attention (ECA) was introduced. With the incorporation of ECA, this model could effectively suppress background noise and provide more focus to insulator regions, thus enhancing its ability to detect insulator defects accurately. Secondly, a partial convolution (PConv) approach was employed in the backbone network instead of conventional convolution, which learned some important channels. This substitution improved both the network model’s accuracy and the training speed. Finally, the Normalized Wasserstein Distance (NWD) prevented insulator features from being lost during pre-feature extraction, which reduced the leakage rate and improved the detection accuracy of small target insulators and defective insulators. The experimental results demonstrate that the improved YOLO v7 network model achieved an average detection accuracy (mAP) of 98.1%, recall of 93.7%, and precision of 96.8% on the TISLTR dataset. On the FISLTR dataset, the average detection accuracy (mAP) for flashover insulators was 93%, with a recall of 92.3% and precision of 87.1%. The average detection accuracy (mAP) for broken insulators was 92.2%, with a recall of 90.3% and a precision of 95.2%. These metrics demonstrate significant improvements in both datasets, highlighting the proposed algorithms’ strong generalization capability and practicable potential to detect insulator targets. Full article

► Show Figures

Figure 1

13 pages, 4943 KiB

Open AccessArticle

Soft Error Simulation of Near-Threshold SRAM Design for Nanosatellite Applications

by Laurent Artola, Benjamin Ruard, Julien Forest and Guillaume Hubert

Electronics 2023, 12(18), 3968; https://doi.org/10.3390/electronics12183968 - 20 Sep 2023

Viewed by 1789

Abstract

This paper presents the benefit of the near-threshold design of random-access memory (SRAM) design to reduce software errors during very low-power operations in nanosatellites. The near-threshold design is based on an optimization of the use of the Schmitt trigger structure for a 45 [...] Read more.

This paper presents the benefit of the near-threshold design of random-access memory (SRAM) design to reduce software errors during very low-power operations in nanosatellites. The near-threshold design is based on an optimization of the use of the Schmitt trigger structure for a 45 nm technology. The results of the soft error susceptibility of the optimized design are compared to a standard 6T SRAM cell. These two designs are modeled and validated by comparing the results with experimental measurements of both static noise margin (SNM) and single event upset (SEU). The optimized circuit reduces the multiple upsets occurrence from 95% down to 14%. Based on the use of simulation tools, the paper demonstrates that the near-threshold design of SRAM is an excellent candidate for the radiation point of view for agile nanosatellites. The results computed for the near-threshold SRAM device demonstrate an improvement of a factor of up to 25 of the soft error rate (SER) in a GEO orbit. Full article

(This article belongs to the Special Issue Advanced CMOS Devices)

► Show Figures

Figure 1

Figure 1
Schematic of (a) basic Schmitt trigger inverter and (b) optimized Schmitt trigger inverter dedicated for 10T ST SRAM cell [<a href="#B3-electronics-12-03968" class="html-bibr">3</a>]. Full article ">Figure 2
Circuits and top views of layouts of SRAM cell: (a) the conventional 6T SRAM cell for 45 nm technologies [<a href="#B6-electronics-12-03968" class="html-bibr">6</a>], (b) the 10T optimized Schmitt trigger SRAM [<a href="#B3-electronics-12-03968" class="html-bibr">3</a>] shrunk to the 45 nm technology. Full article ">Figure 3
(a) Simulated butterfly curves of the 45 nm SRAM in read mode and (b) the simulated SNM (black) and measured SNM (red) [<a href="#B6-electronics-12-03968" class="html-bibr">6</a>] of 6T SRAM cell as a function of core voltage in read mode. Full article ">Figure 4
SRAM cell characteristics modeled for the standard 6T SRAM (black lines) and optimized ST SRAM (red lines) which operate at 1.0 V and 0.3 V. Full article ">Figure 5
Read SNM obtained by simulation for standard 6T SRAM (blue) and the optimized ST SRAM (red) as a function of voltage. Full article ">Figure 6
Hold SNM obtained by simulation for standard 6T SRAM (blue) and the optimized ST SRAM (red) as a function of voltage. Full article ">Figure 7
Evolution of the upset voltage threshold simulated for standard 6T SRAM (black squares) and optimized ST SRAM (red dots) as a function of bias voltage. Full article ">Figure 8
General simulation framework at transistor level based on MUSCA SEP3 and TERRIFIC tools. Full article ">Figure 9
Validation of SEU prediction with experimental data of 45 nm 6T SRAM devices: SPARTAN6 FPGA SRAM based from M. Gadlage et al. [<a href="#B15-electronics-12-03968" class="html-bibr">15</a>] and N. Sukhaseum et al. [<a href="#B16-electronics-12-03968" class="html-bibr">16</a>], and STMicroelectronics SRAMs from G. Gasiot et al. [<a href="#B17-electronics-12-03968" class="html-bibr">17</a>]. Full article ">Figure 10
Estimated SEU cross-section of the standard 6T SRAM in 45 nm technology simulated for heavy ions irradiation as a function of the bias of the memory. Full article ">Figure 11
Enhancement of SEU cross section using ST structure in 45 nm SRAM cells operating at 1 V (empty symbols) and (filled symbols) 0.5 V as a function of the LET of heavy ions. Full article ">Figure 12
Calculated MBU (in red) and SEU (in blue) ratio as a function of the LET of heavy ions (MeV·cm2·mg−1) for 6T SRAM (top) and optimized ST SRAM (bottom) operating at 1 V (left) and 0.4 V (right). Full article ">Figure 13
(a) The 2U nanosatellite platform with its different boards whose the OBC board, (b) potential locations of the OBC board simulated by the tool SEE-U. Full article ">Figure 14
Simulation of the SER mapping as a function of the position of the OBC board in the 2U nanosatellite platform: (a) for the Standard SRAM, (b) for the optimized ST SRAM. Full article ">

25 pages, 21025 KiB

Open AccessArticle

A Multi-Strategy Crazy Sparrow Search Algorithm for the Global Optimization Problem

by Xuewei Jiang, Wei Wang, Yuanyuan Guo and Senlin Liao

Electronics 2023, 12(18), 3967; https://doi.org/10.3390/electronics12183967 - 20 Sep 2023

Cited by 1 | Viewed by 1399

Abstract

A multi-strategy crazy sparrow search algorithm (LTMSSA) for logic-tent hybrid chaotic maps is given in the research to address the issues of poor population diversity, slow convergence, and easily falling into the local optimum of the sparrow search algorithm (SSA). Firstly, the LTMSSA [...] Read more.

A multi-strategy crazy sparrow search algorithm (LTMSSA) for logic-tent hybrid chaotic maps is given in the research to address the issues of poor population diversity, slow convergence, and easily falling into the local optimum of the sparrow search algorithm (SSA). Firstly, the LTMSSA employs an elite chaotic backward learning strategy and an improved discoverer-follower ratio factor to improve the population’s quality and diversity. Secondly, the LTMSSA updates the positions of discoverers and followers by the crazy operator and the Lévy flight strategy to expand the selection range of target following individuals. Finally, during the algorithm’s optimization search, the LTMSSA introduces the tent hybrid and Corsi variable perturbation strategies to improve the population’s ability to jump out of the local optimum. Different types and dimensions of test functions are used as performance benchmark functions to test the performance of the LTMSSA with SSA variants and other algorithms. The simulation results show that the LTMSSA can jump out of the optimal local solution, converge faster, and have higher accuracy. Its overall performance is better than the other seven algorithms, and the LTMSSA can find smaller optimal values than other algorithms in the welded beam and reducer designs. The results confirm that the LTMSSA is an effective aid for computationally complex practical tasks, provides high-quality solutions, and outperforms other algorithms. Full article

(This article belongs to the Special Issue Emerging Technologies and Applications of High-Performance Computer Systems)

► Show Figures

Figure 1

11 pages, 1455 KiB

Open AccessArticle

Enhanced Speech Emotion Recognition Using DCGAN-Based Data Augmentation

by Ji-Young Baek and Seok-Pil Lee

Electronics 2023, 12(18), 3966; https://doi.org/10.3390/electronics12183966 - 20 Sep 2023

Cited by 10 | Viewed by 2764

Abstract

Although emotional speech recognition has received increasing emphasis in research and applications, it remains challenging due to the diversity and complexity of emotions and limited datasets. To address these limitations, we propose a novel approach utilizing DCGAN to augment data from the RAVDESS [...] Read more.

Although emotional speech recognition has received increasing emphasis in research and applications, it remains challenging due to the diversity and complexity of emotions and limited datasets. To address these limitations, we propose a novel approach utilizing DCGAN to augment data from the RAVDESS and EmoDB databases. Then, we assess the efficacy of emotion recognition using mel-spectrogram data by utilizing a model that combines CNN and BiLSTM. The preliminary experimental results reveal that the suggested technique contributes to enhancing the emotional speech identification performance. The results of this study provide directions for further development in the field of emotional speech recognition and the potential for practical applications. Full article

(This article belongs to the Special Issue Theories and Technologies of Network, Data and Information Security)

► Show Figures

Figure 1

25 pages, 564 KiB

Open AccessArticle

Verifiable and Searchable Symmetric Encryption Scheme Based on the Public Key Cryptosystem

by Gangqiang Duan and Shuai Li

Electronics 2023, 12(18), 3965; https://doi.org/10.3390/electronics12183965 - 20 Sep 2023

Cited by 1 | Viewed by 1598

Abstract

With the rapid development of Internet of Things technology and cloud computing technology, all industries need to outsource massive data to third-party clouds for storage in order to reduce storage and computing costs. Verifiable and dynamic searchable symmetric encryption is a very important [...] Read more.

With the rapid development of Internet of Things technology and cloud computing technology, all industries need to outsource massive data to third-party clouds for storage in order to reduce storage and computing costs. Verifiable and dynamic searchable symmetric encryption is a very important cloud security technology, which supports the dynamic update of private data and allows users to perform search operations on the cloud server and verify the legitimacy of the returned results. Therefore, how to realize the dynamic search of encrypted cloud data and the effective verification of the results returned by the cloud server is a key problem to be solved. To solve this problem, we propose a verifiable dynamic encryption scheme (v-PADSSE) based on the public key cryptosystem. In order to achieve efficient and correct data updating, the scheme designs verification information (VI) for each keyword and constructs a verification list (VL) to store it. When dynamic update operations are performed on the cloud data, it is easy to quickly update the security index through obtaining the latest verification information in the VL. The safety and performance evaluation of the v-PADSSE scheme proved that the scheme is safe and effective. Full article

(This article belongs to the Special Issue AI-Driven Network Security and Privacy)

► Show Figures

Figure 1

12 pages, 1402 KiB

Open AccessArticle

A Resource-Efficient Keyword Spotting System Based on a One-Dimensional Binary Convolutional Neural Network

by Jinsung Yoon, Neungyun Kim, Donghyun Lee, Su-Jung Lee, Gil-Ho Kwak and Tae-Hwan Kim

Electronics 2023, 12(18), 3964; https://doi.org/10.3390/electronics12183964 - 20 Sep 2023

Cited by 2 | Viewed by 2015

Abstract

This paper proposes a resource-efficient keyword spotting (KWS) system based on a convolutional neural network (CNN). The end-to-end KWS process is performed based solely on 1D-CNN inference, where features are first extracted from a few convolutional blocks, and then the keywords are classified [...] Read more.

This paper proposes a resource-efficient keyword spotting (KWS) system based on a convolutional neural network (CNN). The end-to-end KWS process is performed based solely on 1D-CNN inference, where features are first extracted from a few convolutional blocks, and then the keywords are classified using a few fully connected blocks. The 1D-CNN model is binarized to reduce resource usage, and its inference is executed by employing a dedicated engine. This engine is designed to skip redundant operations, enabling high inference speed despite its low complexity. The proposed system is implemented using 6895 ALUTs in an Intel Cyclone V FPGA by integrating the essential components for performing the KWS process. In the system, the latency required to process a frame is 22 ms, and the spotting accuracy is 91.80% in an environment where the signal-to-noise ratio is 10 dB for Google speech commands dataset version 2. Full article

(This article belongs to the Section Artificial Intelligence Circuits and Systems (AICAS))

► Show Figures

Figure 1

19 pages, 11266 KiB

Open AccessArticle

Research on Tuning Control Technology for Wireless Power Transfer Systems for Concrete Embedded Sensors

by Cancan Rong, Zhousen Wu, Lihui Yan, Mengmeng Chen, Jiaan Yan, Gang Ren and Chenyang Xia

Electronics 2023, 12(18), 3963; https://doi.org/10.3390/electronics12183963 - 20 Sep 2023

Cited by 2 | Viewed by 1392

Abstract

Concrete embedded sensors play a very important role in structural health monitoring. However, the time of endurance of sensors remains a performance bottleneck and sensors need to be charged without damaging the structure as well. Wireless power transfer (WPT) technology is a promising [...] Read more.

Concrete embedded sensors play a very important role in structural health monitoring. However, the time of endurance of sensors remains a performance bottleneck and sensors need to be charged without damaging the structure as well. Wireless power transfer (WPT) technology is a promising approach to solving this problem. However, the electromagnetic characteristics of concrete medium can cause WPT systems to be untuned and can reduce the energy transmission efficiency of the system. In this paper, the induced medium loss and eddy current loss of a WPT system in concrete are calculated using analytical equations and finite element analysis method. The equivalent circuit model of a concrete–air transmedia WPT system is established according to the calculated losses and a composite tuning control technology is proposed based on the above analysis. In addition, the composite tuning control technology combines the advantages of frequency-modulation tuning and dynamic compensation tuning to ensure the overall resonance of the WPT system. The tuning control technology can ensure the whole resonance of the WPT system and make the natural resonant frequencies of primary and secondary sides consistent. The experimental results show that compared with the untuned control technology, the output power and efficiency of the tuned control system increased by 73% and 11.05%, respectively. The proposed tuning control technology provides direction for future charging of concrete-embedded sensors. Full article

(This article belongs to the Section Power Electronics)

► Show Figures

Figure 1

22 pages, 4842 KiB

Open AccessArticle

Automatic Modulation Classification with Deep Neural Networks

by Clayton A. Harper, Mitchell A. Thornton and Eric C. Larson

Electronics 2023, 12(18), 3962; https://doi.org/10.3390/electronics12183962 - 20 Sep 2023

Cited by 4 | Viewed by 2939

Abstract

Automatic modulation classification is an important component in many modern aeronautical communication systems to achieve efficient spectrum usage in congested wireless environments and other communications systems applications. In recent years, numerous convolutional deep learning architectures have been proposed for automatically classifying the modulation [...] Read more.

Automatic modulation classification is an important component in many modern aeronautical communication systems to achieve efficient spectrum usage in congested wireless environments and other communications systems applications. In recent years, numerous convolutional deep learning architectures have been proposed for automatically classifying the modulation used on observed signal bursts. However, a comprehensive analysis of these differing architectures and the importance of each design element has not been carried out. Thus, it is unclear what trade-offs the differing designs of these convolutional neural networks might have. In this research, we investigate numerous architectures for automatic modulation classification and perform a comprehensive ablation study to investigate the impacts of varying hyperparameters and design elements on automatic modulation classification accuracy. We show that a new state-of-the-art accuracy can be achieved using a subset of the studied design elements, particularly as applied to modulation classification over intercepted bursts of varying time duration. In particular, we show that a combination of dilated convolutions, statistics pooling, and squeeze-and-excitation units results in the strongest performing classifier achieving 98.9% peak accuracy and 63.7% overall accuracy on the RadioML 2018.01A dataset. We further investigate this best performer according to various other criteria, including short signal bursts of varying length, common misclassifications, and performance across differing modulation categories and modes. Full article

(This article belongs to the Topic Machine Learning in Communication Systems and Networks)

► Show Figures

Figure 1

23 pages, 4838 KiB

Open AccessArticle

Mining Highly Visited Co-Location Patterns Based on Minimum Visitor Similarity Constraints

by Xiaoxuan Wang, Peijie Jin, Wen Xiong and Song Gao

Electronics 2023, 12(18), 3961; https://doi.org/10.3390/electronics12183961 - 20 Sep 2023

Viewed by 1164

Abstract

Spatial co-location pattern is a subset of spatial features which shows association relationships based on the spatial neighborhoods. Because the previous prevalence measurements of a co-location pattern have not considered the visited information of spatial instances, co-location patterns do not reflect the social [...] Read more.

Spatial co-location pattern is a subset of spatial features which shows association relationships based on the spatial neighborhoods. Because the previous prevalence measurements of a co-location pattern have not considered the visited information of spatial instances, co-location patterns do not reflect the social connections (such as their spatial instances are constantly visited by common or similar moving objects) between spatial features. In this paper, a special type of co-location pattern, “Highly visited co-location patterns”, is proposed, which considers the spatial proximity and visitor similarity of spatial features at the same time. A new measurement, “Minimum visitor similarity”, has been proposed to reflect the visitor similarity of co-location patterns. By discussing the properties of the minimum visitor similarity, we propose an efficient algorithm to mine the highly visited co-locations and give two pruning strategies to improve the efficiency of the algorithm. Finally, extensive experiments on YELP and Foursquare datasets prove the practicability and efficiency of the proposed algorithm, and we define a “Social Entropy” to prove that spatial features in the co-locations we mined have stronger social connections. Full article

► Show Figures

Figure 1

19 pages, 5943 KiB

Open AccessArticle

A Deep Learning Approach with Extensive Sentiment Analysis for Quantitative Investment

by Wang Li, Chaozhu Hu and Youxi Luo

Electronics 2023, 12(18), 3960; https://doi.org/10.3390/electronics12183960 - 20 Sep 2023

Cited by 2 | Viewed by 2970

Abstract

Recently, deep-learning-based quantitative investment is playing an increasingly important role in the field of finance. However, due to the complexity of the stock market, establishing effective quantitative investment methods is facing challenges from various aspects because of the complexity of the stock market. [...] Read more.

Recently, deep-learning-based quantitative investment is playing an increasingly important role in the field of finance. However, due to the complexity of the stock market, establishing effective quantitative investment methods is facing challenges from various aspects because of the complexity of the stock market. Existing research has inadequately utilized stock news information, overlooking significant details within news content. By constructing a deep hybrid model for comprehensive analysis of historical trading data and news information, complemented by momentum trading strategies, this paper introduces a novel quantitative investment approach. For the first time, we fully consider two dimensions of news, including headlines and contents, and further explore their combined impact on modeling stock price. Our approach initially employs fundamental analysis to screen valuable stocks. Subsequently, we built technical factors based on historical trading data. We then integrated news headlines and content summarized through language models to extract semantic information and representations. Lastly, we constructed a deep neural model to capture global features by combining technical factors with semantic representations, enabling stock prediction and trading decisions. Empirical results conducted on over 4000 stocks from the Chinese stock market demonstrated that incorporating news content enriched semantic information and enhanced objectivity in sentiment analysis. Our proposed method achieved an annualized return rate of 32.06% with a maximum drawdown rate of 5.14%. It significantly outperformed the CSI 300 index, indicating its applicability to guiding investors in making more effective investment strategies and realizing considerable returns. Full article

(This article belongs to the Special Issue Applications of Deep Learning Techniques)

► Show Figures

Figure 1

19 pages, 10975 KiB

Open AccessArticle

The Distributed HTAP Architecture for Real-Time Analysis and Updating of Point Cloud Data

by Juhyun Kim and Changjoo Moon

Electronics 2023, 12(18), 3959; https://doi.org/10.3390/electronics12183959 - 20 Sep 2023

Viewed by 1734

Abstract

Updating the most recent set of point cloud data is critical in autonomous driving environments. However, existing systems for point cloud data management often fail to ensure real-time updates or encounter situations in which data cannot be effectively refreshed. To address these challenges, [...] Read more.

Updating the most recent set of point cloud data is critical in autonomous driving environments. However, existing systems for point cloud data management often fail to ensure real-time updates or encounter situations in which data cannot be effectively refreshed. To address these challenges, this study proposes a distributed hybrid transactional/analytical processing architecture designed for the efficient management and real-time processing of point cloud data. The proposed architecture leverages both columnar and row-based tables, enabling it to handle the substantial workloads associated with its hybrid architecture. The construction of this architecture as a distributed database cluster ensures real-time online analytical process query performance through query parallelization. A dissimilarity analysis algorithm for point cloud data, built by utilizing the capabilities of the spatial database, updates the point cloud data for the relevant area whenever the online analytical process query results indicate high dissimilarity. This research contributes to ensuring real-time hybrid transactional/analytical processing workload processing in dynamic road environments, helping autonomous vehicles generate safe, optimized routes. Full article

(This article belongs to the Special Issue Autonomous Vehicles Technological Trends, 2nd Edition)

► Show Figures

Figure 1

19 pages, 2436 KiB

Open AccessReview

Cybersecurity Risk Analysis in the IoT: A Systematic Review

by Thanaa Saad AlSalem, Mohammed Amin Almaiah and Abdalwali Lutfi

Electronics 2023, 12(18), 3958; https://doi.org/10.3390/electronics12183958 - 20 Sep 2023

Cited by 15 | Viewed by 21469

Abstract

The Internet of Things (IoT) is increasingly becoming a part of our daily lives, raising significant concerns about future cybersecurity risks and the need for reliable solutions. This study conducts a comprehensive systematic literature review to examine the various challenges and attacks threatening [...] Read more.

The Internet of Things (IoT) is increasingly becoming a part of our daily lives, raising significant concerns about future cybersecurity risks and the need for reliable solutions. This study conducts a comprehensive systematic literature review to examine the various challenges and attacks threatening IoT cybersecurity, as well as the proposed frameworks and solutions. Furthermore, it explores emerging trends and identifies existing gaps in this domain. The study’s novelty lies in its extensive exploration of machine learning techniques for detecting and countering IoT threats. It also contributes by highlighting research gaps in economic impact assessment and industrial IoT security. The systematic review analyzes 40 articles, providing valuable insights and guiding future research directions. Results show that privacy issues and cybercrimes are the primary concerns in IoT security, and artificial intelligence holds promise for future cybersecurity. However, some attacks remain inadequately addressed by existing solutions, such as confidentiality, security authentication, and data server connection attacks, necessitating further research and real-life testing of proposed remedies. Full article

(This article belongs to the Special Issue IoT-Enabled Smart Applications for Post-COVID-19)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Electronics, Volume 12, Issue 18 (September-2 2023) – 251 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI