Pub Date : 2024-05-24DOI: 10.3390/electronics13112045
Linyang Zhang, Jianxiang Guo, Xinran Yu, Gang Hui, Na Liu, Dongdong Ren, Jijin Wang
An integrated energy system (IES) plays a key role in transforming energy consumption patterns and solving serious environmental and economic problems. However, the abundant optional schemes and the complex coupling relationship among each piece of equipment make the optimization of an IES very complicated, and most of the current literature focuses on optimization of a specific system. In this work, a simulation-based two-step decoupling method is proposed to simplify the optimization of an IES. The generalized IES is split into four subsystems, and a two-layer optimization method is applied for optimization of the capacity of each piece of equipment. The proposed method enables fast comparison among abundant optional configurations of an IES, and it is applied to a hospital in Beijing, China. The optimized coupling system includes the gas-fired trigeneration system, the GSHP, and the electric chiller. Compared with the traditional distributed systems, the emission reduction rate of CO2 and NOX for the coupling system reaches 153.8% and 314.5%, respectively. Moreover, the primary energy consumption of the coupling system is 82.67% less than that of the traditional distributed energy system, while the annual cost is almost at the same level.
{"title":"Optimization of Integrated Energy Systems Based on Two-Step Decoupling Method","authors":"Linyang Zhang, Jianxiang Guo, Xinran Yu, Gang Hui, Na Liu, Dongdong Ren, Jijin Wang","doi":"10.3390/electronics13112045","DOIUrl":"https://doi.org/10.3390/electronics13112045","url":null,"abstract":"An integrated energy system (IES) plays a key role in transforming energy consumption patterns and solving serious environmental and economic problems. However, the abundant optional schemes and the complex coupling relationship among each piece of equipment make the optimization of an IES very complicated, and most of the current literature focuses on optimization of a specific system. In this work, a simulation-based two-step decoupling method is proposed to simplify the optimization of an IES. The generalized IES is split into four subsystems, and a two-layer optimization method is applied for optimization of the capacity of each piece of equipment. The proposed method enables fast comparison among abundant optional configurations of an IES, and it is applied to a hospital in Beijing, China. The optimized coupling system includes the gas-fired trigeneration system, the GSHP, and the electric chiller. Compared with the traditional distributed systems, the emission reduction rate of CO2 and NOX for the coupling system reaches 153.8% and 314.5%, respectively. Moreover, the primary energy consumption of the coupling system is 82.67% less than that of the traditional distributed energy system, while the annual cost is almost at the same level.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141099611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-24DOI: 10.3390/electronics13112046
Yiru Zhang, Bijing Liu, Yong Yang, Qun Yang
Current target-speaker extraction (TSE) models have achieved good performance in separating target speech from highly overlapped multi-talker speech. However, in real-world applications, multi-talker speech is often sparsely overlapped, and the target speaker may be absent from the speech mixture, making it difficult for the model to extract the desired speech in such situations. To optimize models for various scenarios, universal speaker extraction has been proposed. However, current models do not distinguish between the presence or absence of the target speaker, resulting in suboptimal performance. In this paper, we propose a gated cross-attention network for universal speaker extraction. In our model, the cross-attention mechanism learns the correlation between the target speaker and the speech to determine whether the target speaker is present. Based on this correlation, the gate mechanism enables the model to focus on extracting speech when the target is present and filter out features when the target is absent. Additionally, we propose a joint loss function to evaluate both the reconstructed target speech and silence. Experiments on the WSJ0-2mix-extr and LibriMix datasets show that our proposed method achieves superior performance over comparison approaches in terms of SI-SDR and WER.
当前的目标说话者提取（TSE）模型在从高度重叠的多说话者语音中分离目标语音方面取得了很好的效果。然而，在实际应用中，多说话者语音往往是稀疏重叠的，目标说话者可能不在语音混合物中，因此模型很难在这种情况下提取所需的语音。为了针对各种情况优化模型，有人提出了通用说话人提取方法。然而，目前的模型并不能区分目标说话人的存在与否，从而导致性能不尽如人意。在本文中，我们提出了一种用于通用扬声器提取的门控交叉注意网络。在我们的模型中，交叉注意机制学习目标说话者与语音之间的相关性，以确定目标说话者是否存在。基于这种相关性，门控机制可使模型在目标出现时专注于提取语音，而在目标不存在时过滤掉特征。此外，我们还提出了一个联合损失函数，用于评估重建的目标语音和静音。在 WSJ0-2mix-extr 和 LibriMix 数据集上的实验表明，我们提出的方法在 SI-SDR 和 WER 方面的性能优于对比方法。
{"title":"Gated Cross-Attention for Universal Speaker Extraction: Toward Real-World Applications","authors":"Yiru Zhang, Bijing Liu, Yong Yang, Qun Yang","doi":"10.3390/electronics13112046","DOIUrl":"https://doi.org/10.3390/electronics13112046","url":null,"abstract":"Current target-speaker extraction (TSE) models have achieved good performance in separating target speech from highly overlapped multi-talker speech. However, in real-world applications, multi-talker speech is often sparsely overlapped, and the target speaker may be absent from the speech mixture, making it difficult for the model to extract the desired speech in such situations. To optimize models for various scenarios, universal speaker extraction has been proposed. However, current models do not distinguish between the presence or absence of the target speaker, resulting in suboptimal performance. In this paper, we propose a gated cross-attention network for universal speaker extraction. In our model, the cross-attention mechanism learns the correlation between the target speaker and the speech to determine whether the target speaker is present. Based on this correlation, the gate mechanism enables the model to focus on extracting speech when the target is present and filter out features when the target is absent. Additionally, we propose a joint loss function to evaluate both the reconstructed target speech and silence. Experiments on the WSJ0-2mix-extr and LibriMix datasets show that our proposed method achieves superior performance over comparison approaches in terms of SI-SDR and WER.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141100196","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the rapid development of advanced driver assistance systems (ADASs) and autonomous driving technology, in-vehicle networks are facing huge challenges in real-time operation and data loss. Traditional vehicle bus network systems such as LIN, CAN, and FlexRay are insufficient to meet the real-time requirements of intelligent connected vehicles. In-vehicle Ethernet meets the requirements of high reliability, low electromagnetic radiation, low power consumption, bandwidth allocation, low latency, and real-time synchronization of intelligent connected vehicles. In-vehicle Ethernet has become one of the trends in the next generation of in-vehicle network architecture. This research focuses on the delay problem existing in the real-time data transmission process of in-vehicle Ethernet, and innovatively proposes a fixed point message scheduling algorithm (FPMS) based on time-sensitive network (TSN) technology. By building an experimental platform based on the CANoe simulation tool, the high-efficiency message transmission performance of the fixed point message scheduling algorithm was verified. Experimental results show that the fixed point message scheduling algorithm proposed in this study improves message transmission efficiency by 66%, laying a solid foundation for improving the real-time and reliability performance of in-vehicle Ethernet.
{"title":"Study of Fixed Point Message Scheduling Algorithm for In-Vehicle Ethernet","authors":"Jiaoyue Chen, Qihui Zuo, Yihu Xu, Yujing Wu, Wenquan Jin, Yinan Xu","doi":"10.3390/electronics13112050","DOIUrl":"https://doi.org/10.3390/electronics13112050","url":null,"abstract":"With the rapid development of advanced driver assistance systems (ADASs) and autonomous driving technology, in-vehicle networks are facing huge challenges in real-time operation and data loss. Traditional vehicle bus network systems such as LIN, CAN, and FlexRay are insufficient to meet the real-time requirements of intelligent connected vehicles. In-vehicle Ethernet meets the requirements of high reliability, low electromagnetic radiation, low power consumption, bandwidth allocation, low latency, and real-time synchronization of intelligent connected vehicles. In-vehicle Ethernet has become one of the trends in the next generation of in-vehicle network architecture. This research focuses on the delay problem existing in the real-time data transmission process of in-vehicle Ethernet, and innovatively proposes a fixed point message scheduling algorithm (FPMS) based on time-sensitive network (TSN) technology. By building an experimental platform based on the CANoe simulation tool, the high-efficiency message transmission performance of the fixed point message scheduling algorithm was verified. Experimental results show that the fixed point message scheduling algorithm proposed in this study improves message transmission efficiency by 66%, laying a solid foundation for improving the real-time and reliability performance of in-vehicle Ethernet.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141101139","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-23DOI: 10.3390/electronics13112034
Konstantinos I. Roumeliotis, Nikolaos D. Tselikas, Dimitrios K. Nasiopoulos
Spam emails and phishing attacks continue to pose significant challenges to email users worldwide, necessitating advanced techniques for their efficient detection and classification. In this paper, we address the persistent challenges of spam emails and phishing attacks by introducing a cutting-edge approach to email filtering. Our methodology revolves around harnessing the capabilities of advanced language models, particularly the state-of-the-art GPT-4 Large Language Model (LLM), along with BERT and RoBERTa Natural Language Processing (NLP) models. Through meticulous fine-tuning tailored for spam classification tasks, we aim to surpass the limitations of traditional spam detection systems, such as Convolutional Neural Networks (CNNs). Through an extensive literature review, experimentation, and evaluation, we demonstrate the effectiveness of our approach in accurately identifying spam and phishing emails while minimizing false positives. Our methodology showcases the potential of fine-tuning LLMs for specialized tasks like spam classification, offering enhanced protection against evolving spam and phishing attacks. This research contributes to the advancement of spam filtering techniques and lays the groundwork for robust email security systems in the face of increasingly sophisticated threats.
{"title":"Next-Generation Spam Filtering: Comparative Fine-Tuning of LLMs, NLPs, and CNN Models for Email Spam Classification","authors":"Konstantinos I. Roumeliotis, Nikolaos D. Tselikas, Dimitrios K. Nasiopoulos","doi":"10.3390/electronics13112034","DOIUrl":"https://doi.org/10.3390/electronics13112034","url":null,"abstract":"Spam emails and phishing attacks continue to pose significant challenges to email users worldwide, necessitating advanced techniques for their efficient detection and classification. In this paper, we address the persistent challenges of spam emails and phishing attacks by introducing a cutting-edge approach to email filtering. Our methodology revolves around harnessing the capabilities of advanced language models, particularly the state-of-the-art GPT-4 Large Language Model (LLM), along with BERT and RoBERTa Natural Language Processing (NLP) models. Through meticulous fine-tuning tailored for spam classification tasks, we aim to surpass the limitations of traditional spam detection systems, such as Convolutional Neural Networks (CNNs). Through an extensive literature review, experimentation, and evaluation, we demonstrate the effectiveness of our approach in accurately identifying spam and phishing emails while minimizing false positives. Our methodology showcases the potential of fine-tuning LLMs for specialized tasks like spam classification, offering enhanced protection against evolving spam and phishing attacks. This research contributes to the advancement of spam filtering techniques and lays the groundwork for robust email security systems in the face of increasingly sophisticated threats.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141103126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-23DOI: 10.3390/electronics13112033
Chiang Liang Kok, Chee Kit Ho, Nicholas Tanjodi, Yit Yan Koh
The advent of Internet of Things (IoT) technology has paved the way for innovative solutions in various domains, including aquarium maintenance. An IoT-based automated water changing system emerges as a promising solution to ensure a clean and healthy environment for aquarium inhabitants, thereby alleviating basic chores, particularly for aquarium hobbyists. Conventional solutions often fall short in reliability and affordability, merely focusing on water replacement without addressing other crucial factors. In contrast, this novel system integrates cutting-edge features, leveraging wireless monitoring facilitated by Home Assistant and incorporating water seasoning capabilities. Unlike existing systems, which lack comprehensive monitoring, this solution monitors a plethora of water parameters including water height, pH levels, salinity, temperature, and dissolved solids. This holistic approach enables the system to make informed decisions based on real-time data. Utilizing the gathered data, the system employs advanced algorithms to determine requisite actions. For instance, upon detecting a lower water level, it triggers the water vault to replenish water, ensuring optimal water volume for aquatic life. Additionally, it regulates temperature through heating and cooling mechanisms, ensuring the maintenance of ideal conditions for aquatic organisms. Moreover, the system proactively addresses anomalies by generating indicator requests for parameters beyond its operational scope, thereby facilitating timely intervention by the user. By amalgamating state-of-the-art IoT technology with comprehensive water monitoring and proactive decision making capabilities, this automated water changing system represents a significant advancement in aquarium maintenance, promising enhanced efficiency, reliability, and ultimately, a healthier aquatic ecosystem.
{"title":"A Novel Water Level Control System for Sustainable Aquarium Use","authors":"Chiang Liang Kok, Chee Kit Ho, Nicholas Tanjodi, Yit Yan Koh","doi":"10.3390/electronics13112033","DOIUrl":"https://doi.org/10.3390/electronics13112033","url":null,"abstract":"The advent of Internet of Things (IoT) technology has paved the way for innovative solutions in various domains, including aquarium maintenance. An IoT-based automated water changing system emerges as a promising solution to ensure a clean and healthy environment for aquarium inhabitants, thereby alleviating basic chores, particularly for aquarium hobbyists. Conventional solutions often fall short in reliability and affordability, merely focusing on water replacement without addressing other crucial factors. In contrast, this novel system integrates cutting-edge features, leveraging wireless monitoring facilitated by Home Assistant and incorporating water seasoning capabilities. Unlike existing systems, which lack comprehensive monitoring, this solution monitors a plethora of water parameters including water height, pH levels, salinity, temperature, and dissolved solids. This holistic approach enables the system to make informed decisions based on real-time data. Utilizing the gathered data, the system employs advanced algorithms to determine requisite actions. For instance, upon detecting a lower water level, it triggers the water vault to replenish water, ensuring optimal water volume for aquatic life. Additionally, it regulates temperature through heating and cooling mechanisms, ensuring the maintenance of ideal conditions for aquatic organisms. Moreover, the system proactively addresses anomalies by generating indicator requests for parameters beyond its operational scope, thereby facilitating timely intervention by the user. By amalgamating state-of-the-art IoT technology with comprehensive water monitoring and proactive decision making capabilities, this automated water changing system represents a significant advancement in aquarium maintenance, promising enhanced efficiency, reliability, and ultimately, a healthier aquatic ecosystem.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141104093","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-23DOI: 10.3390/electronics13112037
Donghui Li, Jin Nakazato, Manabu Tsukada
Mobile ad hoc networks are self-organizing networks that do not rely on fixed infrastructure. Smart antennas employ advanced beamforming technology, enabling ultra-long-range directional transmission in wireless networks, which leads to lower power consumption and better utilization of spatial resources. The media access control (MAC) protocol design using smart antennas can lead to efficient usage of channel resources. However, during ultra-long-distance transmissions, there may be significant transport delays. In addition, when using the time division multiple access (TDMA) schemes, it can be difficult to manage conflicts arising from adjacent time slot advancement caused by latency compensation in ultra-long-range propagation. Directional transmission and reception can also cause interference between links that reuse the same time slot. This paper proposes a new distributed dynamic TDMA protocol called State Interaction-based Slot Allocation Protocol (SISAP) to address these issues. This protocol is based on slot states and includes TDMA frame structure, slot allocation process, interference self-avoidance strategy, and slot allocation algorithms. According to the simulation results, the MAC layer design scheme suggested in this paper can achieve ultra-long-distance transmission without conflicts. Additionally, it can reduce the interference between links while space multiplexing. Furthermore, the system exhibits remarkable performance in various network aspects, such as throughput and link delay.
移动特设网络是一种不依赖固定基础设施的自组织网络。智能天线采用先进的波束成形技术，可在无线网络中实现超远距离定向传输，从而降低功耗并更好地利用空间资源。使用智能天线的媒体访问控制（MAC）协议设计可有效利用信道资源。然而，在超长距离传输过程中，可能会出现明显的传输延迟。此外，在使用时分多址（TDMA）方案时，很难管理超远距离传播中延迟补偿引起的相邻时隙提前所产生的冲突。定向发送和接收也会在重复使用相同时隙的链路之间造成干扰。本文提出了一种新的分布式动态 TDMA 协议，称为基于状态交互的时隙分配协议（SISAP），以解决这些问题。该协议基于时隙状态，包括 TDMA 帧结构、时隙分配过程、干扰自规避策略和时隙分配算法。根据仿真结果，本文提出的 MAC 层设计方案可实现无冲突的超长距离传输。此外，它还能在空间复用时减少链路间的干扰。此外，该系统在吞吐量和链路延迟等多个网络方面都表现出卓越的性能。
{"title":"A State-Interactive MAC Layer TDMA Protocol Based on Smart Antennas","authors":"Donghui Li, Jin Nakazato, Manabu Tsukada","doi":"10.3390/electronics13112037","DOIUrl":"https://doi.org/10.3390/electronics13112037","url":null,"abstract":"Mobile ad hoc networks are self-organizing networks that do not rely on fixed infrastructure. Smart antennas employ advanced beamforming technology, enabling ultra-long-range directional transmission in wireless networks, which leads to lower power consumption and better utilization of spatial resources. The media access control (MAC) protocol design using smart antennas can lead to efficient usage of channel resources. However, during ultra-long-distance transmissions, there may be significant transport delays. In addition, when using the time division multiple access (TDMA) schemes, it can be difficult to manage conflicts arising from adjacent time slot advancement caused by latency compensation in ultra-long-range propagation. Directional transmission and reception can also cause interference between links that reuse the same time slot. This paper proposes a new distributed dynamic TDMA protocol called State Interaction-based Slot Allocation Protocol (SISAP) to address these issues. This protocol is based on slot states and includes TDMA frame structure, slot allocation process, interference self-avoidance strategy, and slot allocation algorithms. According to the simulation results, the MAC layer design scheme suggested in this paper can achieve ultra-long-distance transmission without conflicts. Additionally, it can reduce the interference between links while space multiplexing. Furthermore, the system exhibits remarkable performance in various network aspects, such as throughput and link delay.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141106323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-23DOI: 10.3390/electronics13112035
Javad Pourahmadazar, Bal S. Virdee, Tayeb A. Denidni
This study introduces a groundbreaking antenna system, the directive Metasurface Half-Maxwell Fish-Eye (MHMF) lens antenna, tailored specifically for Internet-of-Things (IoT) networks. Designed to operate at 60 GHz, this antenna ingeniously integrates a dipole antenna within a parallel-plate waveguide to illuminate a Half-Maxwell Fish-Eye (HMFE) lens. The HMFE lens serves as a focal point, enabling a crucial high gain for IoT operations. The integration of metasurface structures facilitates the attainment of the gradient refractive index essential for the lens surface. By employing commercial Ansys HFSS software, extensive numerical simulations were conducted to meticulously refine the design, focusing particularly on optimizing the dimensions of unit cells, notably the modified H-shaped cells within the parallel waveguides housing the beam launchers. A functional prototype of the antenna was constructed using a standard PCB manufacturing process. Rigorous testing in an anechoic chamber confirmed the functionality of these manufactured devices, with the experimental results closely aligning with the simulated findings. “Far-field measurements have further confirmed the effectiveness of the antenna, establishing it as a high-gain antenna solution suitable for IoT applications. Specifically, it operates effectively within the 60 GHz range of the electromagnetic spectrum, which is crucial for ensuring reliable communication in IoT devices.” The directive HMFE lens antenna represents a significant advancement in enhancing IoT connectivity and capabilities. Leveraging innovative design concepts and metasurface technology, it heralds a new era of adaptable and efficient IoT systems.
{"title":"Unveiling New IoT Antenna Developments: Planar Multibeam Metasurface Half-Maxwell Fish-Eye Lens with Wavelength Etching","authors":"Javad Pourahmadazar, Bal S. Virdee, Tayeb A. Denidni","doi":"10.3390/electronics13112035","DOIUrl":"https://doi.org/10.3390/electronics13112035","url":null,"abstract":"This study introduces a groundbreaking antenna system, the directive Metasurface Half-Maxwell Fish-Eye (MHMF) lens antenna, tailored specifically for Internet-of-Things (IoT) networks. Designed to operate at 60 GHz, this antenna ingeniously integrates a dipole antenna within a parallel-plate waveguide to illuminate a Half-Maxwell Fish-Eye (HMFE) lens. The HMFE lens serves as a focal point, enabling a crucial high gain for IoT operations. The integration of metasurface structures facilitates the attainment of the gradient refractive index essential for the lens surface. By employing commercial Ansys HFSS software, extensive numerical simulations were conducted to meticulously refine the design, focusing particularly on optimizing the dimensions of unit cells, notably the modified H-shaped cells within the parallel waveguides housing the beam launchers. A functional prototype of the antenna was constructed using a standard PCB manufacturing process. Rigorous testing in an anechoic chamber confirmed the functionality of these manufactured devices, with the experimental results closely aligning with the simulated findings. “Far-field measurements have further confirmed the effectiveness of the antenna, establishing it as a high-gain antenna solution suitable for IoT applications. Specifically, it operates effectively within the 60 GHz range of the electromagnetic spectrum, which is crucial for ensuring reliable communication in IoT devices.” The directive HMFE lens antenna represents a significant advancement in enhancing IoT connectivity and capabilities. Leveraging innovative design concepts and metasurface technology, it heralds a new era of adaptable and efficient IoT systems.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141104664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Cyber–physical systems (CPSs) serve as the pivotal core of Internet of Things (IoT) infrastructures, such as smart grids and intelligent transportation, deploying interconnected sensing devices to monitor operating status. With increasing decentralization, the surge in sensor devices expands the potential vulnerability to cyber attacks. It is imperative to conduct anomaly detection research on the multivariate time series data that these sensors produce to bolster the security of distributed CPSs. However, the high dimensionality, absence of anomaly labels in real-world datasets, and intricate non-linear relationships among sensors present considerable challenges in formulating effective anomaly detection algorithms. Recent deep-learning methods have achieved progress in the field of anomaly detection. Yet, many methods either rely on statistical models that struggle to capture non-linear relationships or use conventional deep learning models like CNN and LSTM, which do not explicitly learn inter-variable correlations. In this study, we propose a novel unsupervised anomaly detection method that integrates Sparse Autoencoder with Graph Transformer network (SGTrans). SGTrans leverages Sparse Autoencoder for the dimensionality reduction and reconstruction of high-dimensional time series, thus extracting meaningful hidden representations. Then, the multivariate time series are mapped into a graph structure. We introduce a multi-head attention mechanism from Transformer into graph structure learning, constructing a Graph Transformer network forecasting module. This module performs attentive information propagation between long-distance sensor nodes and explicitly models the complex temporal dependencies among them to enhance the prediction of future behaviors. Extensive experiments and evaluations on three publicly available real-world datasets demonstrate the effectiveness of our approach.
{"title":"Graph Transformer Network Incorporating Sparse Representation for Multivariate Time Series Anomaly Detection","authors":"Qian Yang, Jiaming Zhang, Junjie Zhang, Cailing Sun, Shanyi Xie, Shangdong Liu, Yimu Ji","doi":"10.3390/electronics13112032","DOIUrl":"https://doi.org/10.3390/electronics13112032","url":null,"abstract":"Cyber–physical systems (CPSs) serve as the pivotal core of Internet of Things (IoT) infrastructures, such as smart grids and intelligent transportation, deploying interconnected sensing devices to monitor operating status. With increasing decentralization, the surge in sensor devices expands the potential vulnerability to cyber attacks. It is imperative to conduct anomaly detection research on the multivariate time series data that these sensors produce to bolster the security of distributed CPSs. However, the high dimensionality, absence of anomaly labels in real-world datasets, and intricate non-linear relationships among sensors present considerable challenges in formulating effective anomaly detection algorithms. Recent deep-learning methods have achieved progress in the field of anomaly detection. Yet, many methods either rely on statistical models that struggle to capture non-linear relationships or use conventional deep learning models like CNN and LSTM, which do not explicitly learn inter-variable correlations. In this study, we propose a novel unsupervised anomaly detection method that integrates Sparse Autoencoder with Graph Transformer network (SGTrans). SGTrans leverages Sparse Autoencoder for the dimensionality reduction and reconstruction of high-dimensional time series, thus extracting meaningful hidden representations. Then, the multivariate time series are mapped into a graph structure. We introduce a multi-head attention mechanism from Transformer into graph structure learning, constructing a Graph Transformer network forecasting module. This module performs attentive information propagation between long-distance sensor nodes and explicitly models the complex temporal dependencies among them to enhance the prediction of future behaviors. Extensive experiments and evaluations on three publicly available real-world datasets demonstrate the effectiveness of our approach.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141103222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Virtual private network (VPN) gateways are widely applied to provide secure end-to-end remote access and to relay reliable interconnected communication in cloud computing. As network convergence nodes, the performance of VPN gateways is limited by traditional methods of packet receiving and sending, the kernel protocol stack and the virtual network interface card. This paper proposes a generic high-performance architecture (GHPA) for VPN gateways in consideration of its generality and performance. In terms of generality, we redesign a generic VPN core framework by modeling a generic VPN communication model, formulating generic VPN core technologies and presenting corresponding core algorithms. In terms of performance, we propose a three-layer GHPA for VPN gateways by designing a VPN packet processing layer based on a data plane development kit (DPDK), implementing a user space basic protocol stack and applying our proposed generic VPN core framework. On the basis of the research work above, we implement a high-performance VPN (HP-VPN) and a traditional VPN (T-VPN) that complies with GHPA and traditional methods, respectively. Experimental results prove that the performance of HP-VPN based on GHPA is superior to T-VPN and other common VPNs in RTT, system throughput, packet forwarding rate and jitter. In addition, GHPA is extensible and applicable for other VPN gateways to improve their performance.
{"title":"A Generic High-Performance Architecture for VPN Gateways","authors":"Chunle Fu, Bailing Wang, Wei Wang, Ruichao Mu, Yunxiao Sun, Guodong Xin, Yongzheng Zhang","doi":"10.3390/electronics13112031","DOIUrl":"https://doi.org/10.3390/electronics13112031","url":null,"abstract":"Virtual private network (VPN) gateways are widely applied to provide secure end-to-end remote access and to relay reliable interconnected communication in cloud computing. As network convergence nodes, the performance of VPN gateways is limited by traditional methods of packet receiving and sending, the kernel protocol stack and the virtual network interface card. This paper proposes a generic high-performance architecture (GHPA) for VPN gateways in consideration of its generality and performance. In terms of generality, we redesign a generic VPN core framework by modeling a generic VPN communication model, formulating generic VPN core technologies and presenting corresponding core algorithms. In terms of performance, we propose a three-layer GHPA for VPN gateways by designing a VPN packet processing layer based on a data plane development kit (DPDK), implementing a user space basic protocol stack and applying our proposed generic VPN core framework. On the basis of the research work above, we implement a high-performance VPN (HP-VPN) and a traditional VPN (T-VPN) that complies with GHPA and traditional methods, respectively. Experimental results prove that the performance of HP-VPN based on GHPA is superior to T-VPN and other common VPNs in RTT, system throughput, packet forwarding rate and jitter. In addition, GHPA is extensible and applicable for other VPN gateways to improve their performance.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141102919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-23DOI: 10.3390/electronics13112039
Jing Zhang, Guangli Wu, Xinlong Bi, Yulong Cui
Video summarization aims to analyze the structure and content of videos and extract key segments to construct summarization that can accurately summarize the main content, allowing users to quickly access the core information without browsing the full video. However, existing methods have difficulties in capturing long-term dependencies when dealing with long videos. On the other hand, there is a large amount of noise in graph structures, which may lead to the influence of redundant information and is not conducive to the effective learning of video features. To solve the above problems, we propose a video summarization generation network based on dynamic graph contrastive learning and feature fusion, which mainly consists of three modules: feature extraction, video encoder, and feature fusion. Firstly, we compute the shot features and construct a dynamic graph by using the shot features as nodes of the graph and the similarity between the shot features as the weights of the edges. In the video encoder, we extract the temporal and structural features in the video using stacked L-G Blocks, where the L-G Block consists of a bidirectional long short-term memory network and a graph convolutional network. Then, the shallow-level features are obtained after processing by L-G Blocks. In order to remove the redundant information in the graph, graph contrastive learning is used to obtain the optimized deep-level features. Finally, to fully exploit the feature information of the video, a feature fusion gate using the gating mechanism is designed to fully fuse the shallow-level features with the deep-level features. Extensive experiments are conducted on two benchmark datasets, TVSum and SumMe, and the experimental results show that our proposed method outperforms most of the current state-of-the-art video summarization methods.
{"title":"Video Summarization Generation Network Based on Dynamic Graph Contrastive Learning and Feature Fusion","authors":"Jing Zhang, Guangli Wu, Xinlong Bi, Yulong Cui","doi":"10.3390/electronics13112039","DOIUrl":"https://doi.org/10.3390/electronics13112039","url":null,"abstract":"Video summarization aims to analyze the structure and content of videos and extract key segments to construct summarization that can accurately summarize the main content, allowing users to quickly access the core information without browsing the full video. However, existing methods have difficulties in capturing long-term dependencies when dealing with long videos. On the other hand, there is a large amount of noise in graph structures, which may lead to the influence of redundant information and is not conducive to the effective learning of video features. To solve the above problems, we propose a video summarization generation network based on dynamic graph contrastive learning and feature fusion, which mainly consists of three modules: feature extraction, video encoder, and feature fusion. Firstly, we compute the shot features and construct a dynamic graph by using the shot features as nodes of the graph and the similarity between the shot features as the weights of the edges. In the video encoder, we extract the temporal and structural features in the video using stacked L-G Blocks, where the L-G Block consists of a bidirectional long short-term memory network and a graph convolutional network. Then, the shallow-level features are obtained after processing by L-G Blocks. In order to remove the redundant information in the graph, graph contrastive learning is used to obtain the optimized deep-level features. Finally, to fully exploit the feature information of the video, a feature fusion gate using the gating mechanism is designed to fully fuse the shallow-level features with the deep-level features. Extensive experiments are conducted on two benchmark datasets, TVSum and SumMe, and the experimental results show that our proposed method outperforms most of the current state-of-the-art video summarization methods.","PeriodicalId":504598,"journal":{"name":"Electronics","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141104014","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：