Pub Date : 2024-12-30DOI: 10.26599/TST.2024.9010051
Fuad A. M. Al-Yarimi;Ramzi Salah;Khaled Mohamoud
This study examines secure and effective data sharing methods for edge computing networks. Traditional methods of sharing data at the edge have issues with security, speed, and consensus. The goal is to develop a Blockchain-based Secure Data Sharing Framework (BSDSF) capable of improving data integrity, latency, and overall network efficiency for edge-cloud computing applications. BSDSF proposes using blockchain technology with Byzantine Fault Tolerance (BFT) and smart contract-based validation as a new method of secure data sharing. It has a two-tiered consensus protocol to meet the needs of edge computing, which requires instantaneous responses. BSDSF employs Byzantine fault tolerance to deal with errors and protect against attacks. Smart contracts automate validation and consensus operations, while edge computing processes data at the attack site. Node validation and failure detection methods monitor network quality and dependability, while system security ensures secure communication between nodes. BSDSF is an important step toward digital freedom and trust by protecting security and improving transaction reliability. The framework demonstrates a reduction in transaction latency by up to 30% and an increase in throughput by 25% compared to traditional edge computing models, positioning BSDSF as a pivotal solution for fostering digital freedom and trust in edge computing environments.
{"title":"Blockchain-Driven Secure Data Sharing Framework for Edge Computing Networks","authors":"Fuad A. M. Al-Yarimi;Ramzi Salah;Khaled Mohamoud","doi":"10.26599/TST.2024.9010051","DOIUrl":"https://doi.org/10.26599/TST.2024.9010051","url":null,"abstract":"This study examines secure and effective data sharing methods for edge computing networks. Traditional methods of sharing data at the edge have issues with security, speed, and consensus. The goal is to develop a Blockchain-based Secure Data Sharing Framework (BSDSF) capable of improving data integrity, latency, and overall network efficiency for edge-cloud computing applications. BSDSF proposes using blockchain technology with Byzantine Fault Tolerance (BFT) and smart contract-based validation as a new method of secure data sharing. It has a two-tiered consensus protocol to meet the needs of edge computing, which requires instantaneous responses. BSDSF employs Byzantine fault tolerance to deal with errors and protect against attacks. Smart contracts automate validation and consensus operations, while edge computing processes data at the attack site. Node validation and failure detection methods monitor network quality and dependability, while system security ensures secure communication between nodes. BSDSF is an important step toward digital freedom and trust by protecting security and improving transaction reliability. The framework demonstrates a reduction in transaction latency by up to 30% and an increase in throughput by 25% compared to traditional edge computing models, positioning BSDSF as a pivotal solution for fostering digital freedom and trust in edge computing environments.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 3","pages":"978-997"},"PeriodicalIF":6.6,"publicationDate":"2024-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10817767","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142905813","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In recent years, Mobile Edge Computing (MEC) has received extensive research attention due to its characteristics, such as real-time data processing and flexible application deployment. However, traditional MEC server deployment relies on the terrestrial Base Stations (BSs), resulting in high deployment costs and limited coverage range. In response to these challenges, air-ground coordination has emerged, which effectively combines the advantages of edge computing and Unmanned Aerial Vehicles (UAVs), providing an effective architecture for edge intelligence. By utilizing the flexibility of UAVs and empowering them into edge nodes with computing resources, the coverage range of MEC can be expanded, thereby reducing the reliance of edge devices on terrestrial BSs. Furthermore, leveraging terrestrial BSs as supplements to the computing power compensates for relatively limited computational capabilities of UAVs. Although extensive studies have been conducted on air-ground coordination, there are few related summaries of application technologies and prospects. Thus, the key technologies of air-ground coordination and applications are comprehensively reviewed in this paper. Finally, to provide guidance for interested researchers, the development trends and potential applications of air-ground coordination are explored.
{"title":"A Review on Air-Ground Coordination in Mobile Edge Computing: Key Technologies, Applications and Future Directions","authors":"Siqi Li;Guoqiang Liu;Li Li;Zhongyuan Zhang;Wenhao Fei;Haolong Xiang","doi":"10.26599/TST.2024.9010142","DOIUrl":"https://doi.org/10.26599/TST.2024.9010142","url":null,"abstract":"In recent years, Mobile Edge Computing (MEC) has received extensive research attention due to its characteristics, such as real-time data processing and flexible application deployment. However, traditional MEC server deployment relies on the terrestrial Base Stations (BSs), resulting in high deployment costs and limited coverage range. In response to these challenges, air-ground coordination has emerged, which effectively combines the advantages of edge computing and Unmanned Aerial Vehicles (UAVs), providing an effective architecture for edge intelligence. By utilizing the flexibility of UAVs and empowering them into edge nodes with computing resources, the coverage range of MEC can be expanded, thereby reducing the reliance of edge devices on terrestrial BSs. Furthermore, leveraging terrestrial BSs as supplements to the computing power compensates for relatively limited computational capabilities of UAVs. Although extensive studies have been conducted on air-ground coordination, there are few related summaries of application technologies and prospects. Thus, the key technologies of air-ground coordination and applications are comprehensively reviewed in this paper. Finally, to provide guidance for interested researchers, the development trends and potential applications of air-ground coordination are explored.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 3","pages":"1359-1386"},"PeriodicalIF":6.6,"publicationDate":"2024-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10817761","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142905817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-30DOI: 10.26599/TST.2024.9010082
Peda Narayana Bathula;M. Sreenivasulu
The importance of secure data sharing in fog computing is increasing due to the growing number of Internet of Things (IoT) devices. This article addresses the privacy and security issues brought up by data sharing in the context of IoT fog computing. The suggested framework, called “BlocFogSec”, secures key management and data sharing through blockchain consensus and smart contracts. Unlike existing solutions, BlocFogSec utilizes two types of smart contracts for secure key exchange and data sharing, while employing a consensus protocol to validate transactions and maintain blockchain integrity. To process and store data effectively at the network edge, the framework makes use of fog computing, notably reducing latency and raising throughput. BlocFogSec successfully blocks unauthorized access and data breaches by restricting transactions to authorized nodes. In addition, the framework uses a consensus protocol to validate and add transactions to the blockchain, guaranteeing data accuracy and immutability. To compare BlocFogSec's performance to that of other models, a number of simulations are conducted. The simulation results indicate that BlocFogSec consistently outperforms existing models, such as Security Services for Fog Computing (SSFC) and Blockchain-based Key Management Scheme (BKMS), in terms of throughput (up to 5135 bytes per second), latency (as low as 7 ms), and resource utilization (70% to 92%). The evaluation also takes into account attack defending accuracy (up to 100%), precision (up to 100%), and recall (up to 99.6%), demonstrating BlocFogSec's effectiveness in identifying and preventing potential attacks.
{"title":"An Integrated Blockchain Framework for Secure Data Sharing in IoT Fog Computing","authors":"Peda Narayana Bathula;M. Sreenivasulu","doi":"10.26599/TST.2024.9010082","DOIUrl":"https://doi.org/10.26599/TST.2024.9010082","url":null,"abstract":"The importance of secure data sharing in fog computing is increasing due to the growing number of Internet of Things (IoT) devices. This article addresses the privacy and security issues brought up by data sharing in the context of IoT fog computing. The suggested framework, called “BlocFogSec”, secures key management and data sharing through blockchain consensus and smart contracts. Unlike existing solutions, BlocFogSec utilizes two types of smart contracts for secure key exchange and data sharing, while employing a consensus protocol to validate transactions and maintain blockchain integrity. To process and store data effectively at the network edge, the framework makes use of fog computing, notably reducing latency and raising throughput. BlocFogSec successfully blocks unauthorized access and data breaches by restricting transactions to authorized nodes. In addition, the framework uses a consensus protocol to validate and add transactions to the blockchain, guaranteeing data accuracy and immutability. To compare BlocFogSec's performance to that of other models, a number of simulations are conducted. The simulation results indicate that BlocFogSec consistently outperforms existing models, such as Security Services for Fog Computing (SSFC) and Blockchain-based Key Management Scheme (BKMS), in terms of throughput (up to 5135 bytes per second), latency (as low as 7 ms), and resource utilization (70% to 92%). The evaluation also takes into account attack defending accuracy (up to 100%), precision (up to 100%), and recall (up to 99.6%), demonstrating BlocFogSec's effectiveness in identifying and preventing potential attacks.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 3","pages":"957-977"},"PeriodicalIF":6.6,"publicationDate":"2024-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10817699","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142905887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-30DOI: 10.26599/TST.2024.9010052
Katta Rajesh Babu;Naramula Venkatesh;K. Shashidhar;Yellampalli Dasaratha Rami Reddy;K. Naga Prakash
In this manuscript, the authors introduce a quantum enabled Reinforcement Algorithm by Universal Features (REMF) as a lightweight solution designed to identify and assess the impact of botnet attacks on 5G Internet of Things (IoT) networks. REMF's primary objective is the swift detection of botnet assaults and their effects, aiming to prevent the initiation of such attacks. The algorithm introduces a novel adaptive classification boosting through reinforcement learning, training on values derived from universal features extracted from network transactions within a given training corpus. During the prediction phase, REMF assesses the Botnet attack confidence of feature values obtained from unlabeled network transactions. It then compares these botnet attack confidence values with the botnet attack confidence of optimal features derived during the training phase to predict the potential impact of the botnet attack, categorizing it as high, moderate, low, or not-an-attack (normal). The performance evaluation results demonstrate that REMF achieves the highest decision accuracy, displaying maximum sensitivity and specificity in predicting the scope of botnet attacks at an early stage. The experimental study illustrates that REMF outperforms existing detection techniques for predicting botnet attacks.
{"title":"An Efficient Quantum Enabled Machine Algorithm by Universal Features for Predicting Botnet Attacks in Digital Twin Enabled IoT Networks","authors":"Katta Rajesh Babu;Naramula Venkatesh;K. Shashidhar;Yellampalli Dasaratha Rami Reddy;K. Naga Prakash","doi":"10.26599/TST.2024.9010052","DOIUrl":"https://doi.org/10.26599/TST.2024.9010052","url":null,"abstract":"In this manuscript, the authors introduce a quantum enabled Reinforcement Algorithm by Universal Features (REMF) as a lightweight solution designed to identify and assess the impact of botnet attacks on 5G Internet of Things (IoT) networks. REMF's primary objective is the swift detection of botnet assaults and their effects, aiming to prevent the initiation of such attacks. The algorithm introduces a novel adaptive classification boosting through reinforcement learning, training on values derived from universal features extracted from network transactions within a given training corpus. During the prediction phase, REMF assesses the Botnet attack confidence of feature values obtained from unlabeled network transactions. It then compares these botnet attack confidence values with the botnet attack confidence of optimal features derived during the training phase to predict the potential impact of the botnet attack, categorizing it as high, moderate, low, or not-an-attack (normal). The performance evaluation results demonstrate that REMF achieves the highest decision accuracy, displaying maximum sensitivity and specificity in predicting the scope of botnet attacks at an early stage. The experimental study illustrates that REMF outperforms existing detection techniques for predicting botnet attacks.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 3","pages":"947-956"},"PeriodicalIF":6.6,"publicationDate":"2024-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10817765","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142905888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-30DOI: 10.26599/TST.2024.9010026
Boyuan Yan;Yankun Zhang;Wenwen Gong;Haoyang Wan;Wenwei Wang;Weiyi Zhong;Caixia Bu
Developers integrate web Application Programming Interfaces (APIs) into edge applications, enabling data expansion to the edge computing area for comprehensive coverage of devices in that region. To develop edge applications, developers search API categories to select APIs that meet specific functionalities. Therefore, the accurate classification of APIs becomes critically important. However, existing approaches, as evident on platforms like programableweb.com, face significant challenges. Firstly, sparsity in API data reduces classification accuracy in works focusing on single-dimensional API information. Secondly, the multidimensional and heterogeneous structure of web APIs adds complexity to data mining tasks, requiring sophisticated techniques for effective integration and analysis of diverse data aspects. Lastly, the long-tailed distribution of API data introduces biases, compromising the fairness of classification efforts. Addressing these challenges, we propose MDGCN-Lt, an API classification approach offering flexibility in using multi-dimensional heterogeneous data. It tackles data sparsity through deep graph convolutional networks, exploring high-order feature interactions among API nodes. MDGCN-Lt employs a loss function with logit adjustment, enhancing efficiency in handling long-tail data scenarios. Empirical results affirm our approach's superiority over existing methods.
{"title":"MDGCN-Lt: Fair Web API Classification with Sparse and Heterogeneous Data Based on Deep GCN","authors":"Boyuan Yan;Yankun Zhang;Wenwen Gong;Haoyang Wan;Wenwei Wang;Weiyi Zhong;Caixia Bu","doi":"10.26599/TST.2024.9010026","DOIUrl":"https://doi.org/10.26599/TST.2024.9010026","url":null,"abstract":"Developers integrate web Application Programming Interfaces (APIs) into edge applications, enabling data expansion to the edge computing area for comprehensive coverage of devices in that region. To develop edge applications, developers search API categories to select APIs that meet specific functionalities. Therefore, the accurate classification of APIs becomes critically important. However, existing approaches, as evident on platforms like programableweb.com, face significant challenges. Firstly, sparsity in API data reduces classification accuracy in works focusing on single-dimensional API information. Secondly, the multidimensional and heterogeneous structure of web APIs adds complexity to data mining tasks, requiring sophisticated techniques for effective integration and analysis of diverse data aspects. Lastly, the long-tailed distribution of API data introduces biases, compromising the fairness of classification efforts. Addressing these challenges, we propose MDGCN-Lt, an API classification approach offering flexibility in using multi-dimensional heterogeneous data. It tackles data sparsity through deep graph convolutional networks, exploring high-order feature interactions among API nodes. MDGCN-Lt employs a loss function with logit adjustment, enhancing efficiency in handling long-tail data scenarios. Empirical results affirm our approach's superiority over existing methods.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 3","pages":"1294-1314"},"PeriodicalIF":6.6,"publicationDate":"2024-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10817770","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142905808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As individuals age, their facial features change, which can hinder the accuracy of face recognition technology. To address this challenge, a new cross-age face recognition algorithm, leveraging deep learning and a loss function (Loss), has been proposed in this article. The Retinaface algorithm detects faces in images, while the Resnet-50 model is enhanced by incorporating an attention mechanism and improved softmax loss (Arcface) to extract facial features. This approach has been tested on publicly available and custom-built datasets, and its performance has been compared to other cross-age face recognition techniques. The results show that the model effectively recognizes faces across different age groups.
随着个人年龄的增长，他们的面部特征会发生变化，这可能会影响人脸识别技术的准确性。为了应对这一挑战，本文提出了一种新的跨年龄人脸识别算法，该算法利用深度学习和损失函数（loss）。retaface算法在图像中检测人脸，而Resnet-50模型通过纳入注意机制和改进的softmax loss （Arcface）来提取面部特征。该方法已在公开可用和定制的数据集上进行了测试，并将其性能与其他跨年龄人脸识别技术进行了比较。结果表明，该模型能有效识别不同年龄段的人脸。
{"title":"DEFOG: Deep Learning with Attention Mechanism Enabled Cross-Age Face Recognition","authors":"Biaokai Zhu;Lu Li;Xiaochun Hu;Fulin Wu;Zhaojie Zhang;Shengnan Zhu;Yanxi Wang;Jiali Wu;Jie Song;Feng Li;Sanman Liu;Jumin Zhao","doi":"10.26599/TST.2024.9010107","DOIUrl":"https://doi.org/10.26599/TST.2024.9010107","url":null,"abstract":"As individuals age, their facial features change, which can hinder the accuracy of face recognition technology. To address this challenge, a new cross-age face recognition algorithm, leveraging deep learning and a loss function (Loss), has been proposed in this article. The Retinaface algorithm detects faces in images, while the Resnet-50 model is enhanced by incorporating an attention mechanism and improved softmax loss (Arcface) to extract facial features. This approach has been tested on publicly available and custom-built datasets, and its performance has been compared to other cross-age face recognition techniques. The results show that the model effectively recognizes faces across different age groups.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 3","pages":"1342-1358"},"PeriodicalIF":6.6,"publicationDate":"2024-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10817764","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142905819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-09DOI: 10.26599/TST.2023.9010128
Meiling Li;Peng Xue;Hu Yuan;Yuxing Han
Cooperative jamming can effectively combat eavesdropping in physical layer security communication without affecting the legal receiver and improve the security performance of the system. This paper introduces cooperative jamming to cognitive radio (CR) networks with non-orthogonal multiple access (NOMA) technology. The secure performance of the considered CR and NOMA (CR-NOMA) network is evaluated using two modes: non-cooperative jamming and cooperative jamming. In particular, the secrecy outage probabilities (SOPs) of the primary user (PU) in the two modes are derived under Rician fading channels, based on which, the influences of the transmission signal-to-noise ratio (SNR) of secondary users (SUs), the number of SUs, the secrecy rate, and the power allocation coefficient on the SOPs of PU are analyzed thereafter. Both analysis and simulation results show that cooperative jamming effectively prevents eavesdropping behaviour, which reduces the SOP of PU compared to non-cooperative jamming. We also show that the transmission SNR, the number of SUs, the secrecy rate, and the power distribution coefficients greatly influence performance improvement.
{"title":"Physical Layer Security for CR-NOMA Network with Cooperative Jamming","authors":"Meiling Li;Peng Xue;Hu Yuan;Yuxing Han","doi":"10.26599/TST.2023.9010128","DOIUrl":"https://doi.org/10.26599/TST.2023.9010128","url":null,"abstract":"Cooperative jamming can effectively combat eavesdropping in physical layer security communication without affecting the legal receiver and improve the security performance of the system. This paper introduces cooperative jamming to cognitive radio (CR) networks with non-orthogonal multiple access (NOMA) technology. The secure performance of the considered CR and NOMA (CR-NOMA) network is evaluated using two modes: non-cooperative jamming and cooperative jamming. In particular, the secrecy outage probabilities (SOPs) of the primary user (PU) in the two modes are derived under Rician fading channels, based on which, the influences of the transmission signal-to-noise ratio (SNR) of secondary users (SUs), the number of SUs, the secrecy rate, and the power allocation coefficient on the SOPs of PU are analyzed thereafter. Both analysis and simulation results show that cooperative jamming effectively prevents eavesdropping behaviour, which reduces the SOP of PU compared to non-cooperative jamming. We also show that the transmission SNR, the number of SUs, the secrecy rate, and the power distribution coefficients greatly influence performance improvement.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"708-720"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786938","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Accurate and portable Blood Pressure (BP) monitoring is vital for managing cardiovascular diseases. However, existing wearable continuous BP monitoring technologies are often inaccurate and rely on external calibration, limiting their practical application in continuous BP monitoring. To address this challenge, we have developed a Wearable continuous non-invasive BP Monitor (WeBPM) equipped with a finger cuff sensor, capable of monitoring BP continuously and accurately within medical-grade precision. WeBPM integrates advanced finger oscillographic BP measurement technology to provide reliable self-calibration functionality. Moreover, Pulsatile Cycle Volume Adjustment Method (PCVAM) we proposed for the closed-loop phase can continuously track changes in vasomotor tone under a controlled frequency based on pulsatile cycles, thereby enabling continuous BP measurement. In comparative experiments with the Nexfin monitor, WeBPM demonstrates excellent performance in induced dynamic BP experiments, with measurement errors of (-1.4 ± 6.24) mmHg for Systolic BP (SBP) and (-0.82 ± 4.83) mmHg for Diastolic BP (DBP). Additionally, compared to clinical invasive reference measurements, WeBPM's SBP and DBP measurement errors are (-1.74 ± 4.9) mmHg and (0.37 ± 3.28) mmHg, respectively, further proving its outstanding performance. These results highlight WeBPM's potential in personalized health management and remote monitoring, offering a new solution for continuous non-invasive BP monitoring.
{"title":"Wearable Continuous Blood Pressure Monitoring Based on Pulsatile Cycle Volume Adjustment Method","authors":"Pang Wu;Zhongrui Bai;Pan Xia;Lirui Xu;Peng Wang;Xianxiang Chen;Lidong Du;Ziqing Hei;Weifeng Yao;Xiaoran Li;Zhan Zhao;Zhen Fang","doi":"10.26599/TST.2024.9010043","DOIUrl":"https://doi.org/10.26599/TST.2024.9010043","url":null,"abstract":"Accurate and portable Blood Pressure (BP) monitoring is vital for managing cardiovascular diseases. However, existing wearable continuous BP monitoring technologies are often inaccurate and rely on external calibration, limiting their practical application in continuous BP monitoring. To address this challenge, we have developed a Wearable continuous non-invasive BP Monitor (WeBPM) equipped with a finger cuff sensor, capable of monitoring BP continuously and accurately within medical-grade precision. WeBPM integrates advanced finger oscillographic BP measurement technology to provide reliable self-calibration functionality. Moreover, Pulsatile Cycle Volume Adjustment Method (PCVAM) we proposed for the closed-loop phase can continuously track changes in vasomotor tone under a controlled frequency based on pulsatile cycles, thereby enabling continuous BP measurement. In comparative experiments with the Nexfin monitor, WeBPM demonstrates excellent performance in induced dynamic BP experiments, with measurement errors of (-1.4 ± 6.24) mmHg for Systolic BP (SBP) and (-0.82 ± 4.83) mmHg for Diastolic BP (DBP). Additionally, compared to clinical invasive reference measurements, WeBPM's SBP and DBP measurement errors are (-1.74 ± 4.9) mmHg and (0.37 ± 3.28) mmHg, respectively, further proving its outstanding performance. These results highlight WeBPM's potential in personalized health management and remote monitoring, offering a new solution for continuous non-invasive BP monitoring.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"650-669"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786939","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825881","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-09DOI: 10.26599/TST.2024.9010009
Hongru Zhou;Shengxin Liu;Ruidi Cao
Cohesive subgraph search is a fundamental problem in bipartite graph analysis. Given integers $k$ and ℓ, a (k,ℓ)-biplex is a cohesive structure which requires each vertex to disconnect at most $k$ or $l$ vertices in the other side. Computing (k,ℓ)-biplexes has been a popular research topic in recent years and has various applications. However, most existing studies considered the problem of finding (k, ℓ)-biplex with the largest number of edges. In this paper, we instead consider another variant and focus on the maximum vertex (k, ℓ)-biplex problem which aims to search for a (k, ℓ)-biplex with the maximum cardinality. We first show that this problem is Non-deterministic Polynomial-time hard (NP-hard) for any positive integers $k$ and ℓ while max{k, ℓ} is at least 3. Guided by this negative result, we design an efficient branch-and-bound algorithm with a novel framework. In particular, we introduce a branching strategy based on whether there is a pivot in the current set, with which our proposed algorithm has the time complexity of γ n n O(1) , where γ< 2. In addition, we also apply multiple speed-up techniques and various pruning strategies. Finally, we conduct extensive experiments on various real datasets which demonstrate the efficiency of our proposed algorithm in terms of running time.
{"title":"Efficient Maximum Vertex (k,ℓ)-Biplex Computation on Bipartite Graphs","authors":"Hongru Zhou;Shengxin Liu;Ruidi Cao","doi":"10.26599/TST.2024.9010009","DOIUrl":"https://doi.org/10.26599/TST.2024.9010009","url":null,"abstract":"Cohesive subgraph search is a fundamental problem in bipartite graph analysis. Given integers \u0000<tex>$k$</tex>\u0000 and ℓ, a (k,ℓ)-biplex is a cohesive structure which requires each vertex to disconnect at most \u0000<tex>$k$</tex>\u0000 or \u0000<tex>$l$</tex>\u0000 vertices in the other side. Computing (k,ℓ)-biplexes has been a popular research topic in recent years and has various applications. However, most existing studies considered the problem of finding (k, ℓ)-biplex with the largest number of edges. In this paper, we instead consider another variant and focus on the maximum vertex (k, ℓ)-biplex problem which aims to search for a (k, ℓ)-biplex with the maximum cardinality. We first show that this problem is Non-deterministic Polynomial-time hard (NP-hard) for any positive integers \u0000<tex>$k$</tex>\u0000 and ℓ while max{k, ℓ} is at least 3. Guided by this negative result, we design an efficient branch-and-bound algorithm with a novel framework. In particular, we introduce a branching strategy based on whether there is a pivot in the current set, with which our proposed algorithm has the time complexity of γ\u0000<sup>n</sup>\u0000n\u0000<sup>O(1)</sup>\u0000, where γ< 2. In addition, we also apply multiple speed-up techniques and various pruning strategies. Finally, we conduct extensive experiments on various real datasets which demonstrate the efficiency of our proposed algorithm in terms of running time.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"569-584"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786929","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-09DOI: 10.26599/TST.2023.9010134
Bokai Yang;Hongyang Lei;Huazhen Huang;Xinxin Han;Yunpeng Cai
Radiology report generation is of significant importance. Unlike standard image captioning tasks, radiology report generation faces more pronounced visual and textual biases due to constrained data availability, making it increasingly reliant on prior knowledge in this context. In this paper, we introduce a radiology report generation network termed Dynamics Priori Networks (DPN), which leverages a dynamic knowledge graph and prior knowledge. Concretely, we establish an adaptable graph network and harness both medical domain knowledge and expert insights to enhance the model's intelligence. Notably, we introduce an image-text contrastive module and an image-text matching module to enhance the quality of the generated results. Our method is evaluated on two widely available datasets: X-ray collection from Indiana University (IU X-ray) and Medical Information Mart for Intensive Care, Chest X-Ray (MIMIC-CXR), where it demonstrates superior performance, particularly excelling in critical metrics.
放射学报告生成非常重要。与标准的图像标题任务不同，由于数据可用性的限制，放射学报告生成面临着更明显的视觉和文本偏差，因此在这种情况下越来越依赖于先验知识。在本文中，我们介绍了一种称为动态先验网络（DPN）的放射学报告生成网络，它利用了动态知识图谱和先验知识。具体来说，我们建立了一个可适应的图网络，并利用医学领域知识和专家见解来增强模型的智能性。值得注意的是，我们引入了图像-文本对比模块和图像-文本匹配模块，以提高生成结果的质量。我们的方法在两个广泛可用的数据集上进行了评估：我们的方法在两个广泛使用的数据集上进行了评估：印第安纳大学的 X 射线集（IU X-ray）和重症监护医学信息中心的胸部 X 射线集（MIMIC-CXR），在这两个数据集上，我们的方法表现出了卓越的性能，尤其是在关键指标方面。
{"title":"DPN: Dynamics Priori Networks for Radiology Report Generation","authors":"Bokai Yang;Hongyang Lei;Huazhen Huang;Xinxin Han;Yunpeng Cai","doi":"10.26599/TST.2023.9010134","DOIUrl":"https://doi.org/10.26599/TST.2023.9010134","url":null,"abstract":"Radiology report generation is of significant importance. Unlike standard image captioning tasks, radiology report generation faces more pronounced visual and textual biases due to constrained data availability, making it increasingly reliant on prior knowledge in this context. In this paper, we introduce a radiology report generation network termed Dynamics Priori Networks (DPN), which leverages a dynamic knowledge graph and prior knowledge. Concretely, we establish an adaptable graph network and harness both medical domain knowledge and expert insights to enhance the model's intelligence. Notably, we introduce an image-text contrastive module and an image-text matching module to enhance the quality of the generated results. Our method is evaluated on two widely available datasets: X-ray collection from Indiana University (IU X-ray) and Medical Information Mart for Intensive Care, Chest X-Ray (MIMIC-CXR), where it demonstrates superior performance, particularly excelling in critical metrics.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"600-609"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786932","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825974","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","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：