The widespread implementation of Artificial Itelligence (AI) has led to significant advancements in disease diagnosis. Personalized Federated Learning (FL) trains models tailored to each patient’s needs but often overlooks model architecture heterogeneity. We propose a novel Co-training-based personalized FL with Generative Adversarial Networks (GANs) for Smart Healthcare Diagnosis (CFG-SHD). This approach allows privacy-preserving participation in FL by enabling patients to keep their model architectures and parameters private. Key contributions include integrating co-training into FL for leveraging multiple data views and using GANs to generate synthetic data, ensuring data privacy. By addressing model architecture heterogeneity our approach offers a robust solution for personalized healthcare diagnostics, aligning with the diverse needs of modern healthcare systems and advancing patient-centric AI applications. CFG-SHD enhances personalized diagnosis accuracy, achieving 97.16%, 98.04%, and 97.88% on the PAD-UFES-20, HAM10000, and PH2 datasets, respectively.
人工智能（AI）的广泛应用使疾病诊断取得了重大进展。个性化联邦学习（FL）根据每个患者的需求训练模型，但往往忽略了模型架构的异质性。我们提出了一种基于生成对抗网络（GANs）的基于协同训练的个性化FL，用于智能医疗诊断（CFG-SHD）。这种方法通过使患者保持其模型架构和参数的私密性，从而允许在FL中保护隐私。主要贡献包括将协同训练集成到FL中，以利用多个数据视图，并使用gan生成合成数据，确保数据隐私。通过解决模型架构的异质性，我们的方法为个性化医疗诊断提供了一个强大的解决方案，与现代医疗系统的多样化需求保持一致，并推进以患者为中心的人工智能应用程序。CFG-SHD提高了个性化诊断的准确性，在pad - upes -20、HAM10000和PH2数据集上分别达到97.16%、98.04%和97.88%。
{"title":"Co-Training-Based Personalized Federated Learning With Generative Adversarial Networks for Enhanced Mobile Smart Healthcare Diagnosis","authors":"Arikumar K. Selvaraj;Sahaya Beni Prathiba;A. Deepak Kumar;R. Dhanalakshmi;Thippa Reddy Gadekallu;Gautam Srivastava","doi":"10.1109/TCE.2024.3460469","DOIUrl":"10.1109/TCE.2024.3460469","url":null,"abstract":"The widespread implementation of Artificial Itelligence (AI) has led to significant advancements in disease diagnosis. Personalized Federated Learning (FL) trains models tailored to each patient’s needs but often overlooks model architecture heterogeneity. We propose a novel Co-training-based personalized FL with Generative Adversarial Networks (GANs) for Smart Healthcare Diagnosis (CFG-SHD). This approach allows privacy-preserving participation in FL by enabling patients to keep their model architectures and parameters private. Key contributions include integrating co-training into FL for leveraging multiple data views and using GANs to generate synthetic data, ensuring data privacy. By addressing model architecture heterogeneity our approach offers a robust solution for personalized healthcare diagnostics, aligning with the diverse needs of modern healthcare systems and advancing patient-centric AI applications. CFG-SHD enhances personalized diagnosis accuracy, achieving 97.16%, 98.04%, and 97.88% on the PAD-UFES-20, HAM10000, and PH2 datasets, respectively.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"70 3","pages":"6131-6139"},"PeriodicalIF":4.3,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142258684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-12DOI: 10.1109/TCE.2024.3458985
Zakaria Abou El Houda;Hajar Moudoud;Bouziane Brik;Muhammad Adil
The proliferation of consumer networks has increased vulnerabilities to network intrusions, emphasizing the critical need for robust intrusion detection systems (IDS). The data-driven Artificial Intelligence (AI) approach has gained attention for enhancing IDS capabilities to deal with emerging security threats. However, these AI-based IDS face challenges in scalability and privacy preservation. More importantly, they are time-consuming and may perform poorly on high-dimensional and complex data due to the lack of computational resources. To address these shortcomings, in this paper, we introduce a novel framework, called Quantum Federated Learning IDS (QFL-IDS), that merges Quantum Computing (QC) with Federated Learning (FL) to allow for an efficient, robust, and privacy-preserving approach for detecting network intrusions in consumer networks. Leveraging the decentralized nature of FL, QFL-IDS enables multiple consumer devices to collaboratively train a global intrusion detection model while preserving the privacy of individual user data. Furthermore, we leverage the computational power of quantum computing to improve the efficiency of model training and inference processes. We demonstrate the efficacy of our framework through extensive experiments. The obtained results show significant improvements in detection accuracy and computational efficiency compared to the current traditional centralized and federated learning approaches. This makes QFL-IDS a promising framework to cope with the new emerging security threats in a timely and effective manner.
{"title":"A Privacy-Preserving Framework for Efficient Network Intrusion Detection in Consumer Network Using Quantum Federated Learning","authors":"Zakaria Abou El Houda;Hajar Moudoud;Bouziane Brik;Muhammad Adil","doi":"10.1109/TCE.2024.3458985","DOIUrl":"10.1109/TCE.2024.3458985","url":null,"abstract":"The proliferation of consumer networks has increased vulnerabilities to network intrusions, emphasizing the critical need for robust intrusion detection systems (IDS). The data-driven Artificial Intelligence (AI) approach has gained attention for enhancing IDS capabilities to deal with emerging security threats. However, these AI-based IDS face challenges in scalability and privacy preservation. More importantly, they are time-consuming and may perform poorly on high-dimensional and complex data due to the lack of computational resources. To address these shortcomings, in this paper, we introduce a novel framework, called Quantum Federated Learning IDS (QFL-IDS), that merges Quantum Computing (QC) with Federated Learning (FL) to allow for an efficient, robust, and privacy-preserving approach for detecting network intrusions in consumer networks. Leveraging the decentralized nature of FL, QFL-IDS enables multiple consumer devices to collaboratively train a global intrusion detection model while preserving the privacy of individual user data. Furthermore, we leverage the computational power of quantum computing to improve the efficiency of model training and inference processes. We demonstrate the efficacy of our framework through extensive experiments. The obtained results show significant improvements in detection accuracy and computational efficiency compared to the current traditional centralized and federated learning approaches. This makes QFL-IDS a promising framework to cope with the new emerging security threats in a timely and effective manner.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"70 4","pages":"7121-7128"},"PeriodicalIF":4.3,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142189659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Cloud computing is an emerging choice among businesses all over the world since it provides flexible and world wide Web computer capabilities as a customizable service. Because of the dispersed nature of cloud services, security is a major problem. Since it is extremely accessible to intruders for any kind of assault, privacy and security are major hurdles to the on-demand service’s success. A massive increase in network traffic has opened the path for increasingly difficult and broad security vulnerabilities. The use of traditional Intrusion Detection Systems (IDS) to prevent these attempts has proven ineffective. Therefore, this paper proposes a novel Network Intrusion Detection System (NIDS) based on a Machine Learning (ML) model known as the Support Vector Machine (SVM) and eXtreme Gradient Boosting (XGBoost) techniques. Furthermore, the hyperparameter optimization technique based on the Crow Search Algorithm is being utilized to optimize the NIDS’ performance. Besides, the XGBoost-based feature selection technique is used to improve the classification accuracy of NIDS’s method. Finally, the performance of the proposed system is evaluated using the NSL-KDD and UNR-IDD datasets, and the experiment results show that it performs better than baselines and has the potential to be used in modern NIDS.
{"title":"Machine Learning-Based Network Intrusion Detection Optimization for Cloud Computing Environments","authors":"Jitendra Kumar Samriya;Surendra Kumar;Mohit Kumar;Huaming Wu;Sukhpal Singh Gill","doi":"10.1109/TCE.2024.3458810","DOIUrl":"10.1109/TCE.2024.3458810","url":null,"abstract":"Cloud computing is an emerging choice among businesses all over the world since it provides flexible and world wide Web computer capabilities as a customizable service. Because of the dispersed nature of cloud services, security is a major problem. Since it is extremely accessible to intruders for any kind of assault, privacy and security are major hurdles to the on-demand service’s success. A massive increase in network traffic has opened the path for increasingly difficult and broad security vulnerabilities. The use of traditional Intrusion Detection Systems (IDS) to prevent these attempts has proven ineffective. Therefore, this paper proposes a novel Network Intrusion Detection System (NIDS) based on a Machine Learning (ML) model known as the Support Vector Machine (SVM) and eXtreme Gradient Boosting (XGBoost) techniques. Furthermore, the hyperparameter optimization technique based on the Crow Search Algorithm is being utilized to optimize the NIDS’ performance. Besides, the XGBoost-based feature selection technique is used to improve the classification accuracy of NIDS’s method. Finally, the performance of the proposed system is evaluated using the NSL-KDD and UNR-IDD datasets, and the experiment results show that it performs better than baselines and has the potential to be used in modern NIDS.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"70 4","pages":"7449-7460"},"PeriodicalIF":4.3,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142189662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Process mining, a burgeoning discipline within data science, demonstrates a significant contribution to the software development lifecycle of diverse real-time consumer-centric projects. This paper underscores the prominence of integrating predictive business process monitoring into organizational process models, as it can substantially impact profits and efficiency in any possible business domain along with improving services to consumers. The paper proposes a novel deep learning-based business process prediction model consisting of multiple layers with fine-tuning hyperparameters. The proposed model leverages input embeddings to represent each of the activities, and based on the training of the proposed model, the accuracy of the next activity is calculated. To assess the efficacy of the proposed model, it has been compared with the existing benchmark models. Our proposed model has shown a significant gain over the existing approaches. The results show that the proposed model outperforms these approaches by achieving an accuracy of 76% on the consumer helpdesk dataset along with an accuracy of 78% on the benchmark BPI dataset.
{"title":"Predictive Monitoring in Process Mining Using Deep Learning for Better Consumer Service","authors":"Vasanth Yarlagadda;Abishi Chowdhury;Amrit Pal;Shruti Mishra;Sandeep Kumar Satapathy;Sung-Bae Cho;Sachi Nandan Mohanty;Ashit Kumar Dutta","doi":"10.1109/TCE.2024.3456677","DOIUrl":"10.1109/TCE.2024.3456677","url":null,"abstract":"Process mining, a burgeoning discipline within data science, demonstrates a significant contribution to the software development lifecycle of diverse real-time consumer-centric projects. This paper underscores the prominence of integrating predictive business process monitoring into organizational process models, as it can substantially impact profits and efficiency in any possible business domain along with improving services to consumers. The paper proposes a novel deep learning-based business process prediction model consisting of multiple layers with fine-tuning hyperparameters. The proposed model leverages input embeddings to represent each of the activities, and based on the training of the proposed model, the accuracy of the next activity is calculated. To assess the efficacy of the proposed model, it has been compared with the existing benchmark models. Our proposed model has shown a significant gain over the existing approaches. The results show that the proposed model outperforms these approaches by achieving an accuracy of 76% on the consumer helpdesk dataset along with an accuracy of 78% on the benchmark BPI dataset.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"70 4","pages":"7279-7290"},"PeriodicalIF":4.3,"publicationDate":"2024-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142189660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-05DOI: 10.1109/tce.2024.3454270
Huiru Yan, Yan Gu, Haoyang He, Xin Ning, Qingle Wang, Long Cheng
{"title":"DNN-Based Task Partitioning and Offloading in Edge-Cloud Collaboration Within Electric Vehicles","authors":"Huiru Yan, Yan Gu, Haoyang He, Xin Ning, Qingle Wang, Long Cheng","doi":"10.1109/tce.2024.3454270","DOIUrl":"https://doi.org/10.1109/tce.2024.3454270","url":null,"abstract":"","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"23 1","pages":""},"PeriodicalIF":4.3,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142189661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-03DOI: 10.1109/TCE.2024.3454178
Yanrui Wang;Yue Xiao;Ming Xiao;Nan Li
A family of offset spatial modulation (OSM) and offset space shift keying (OSSK) techniques has been proposed in (Fang et al., 2019) due to their alleviated requirements for radio frequency (RF) switching, toward an efficient design for low-cost consumer electronic (CE) devices with multiple antennas. Yet, the original structure of OSM/OSSK is based on multiple-input single-output (MISO) design, and so far there is no efficient solution to bridge such system with multiple-input multiple-output (MIMO) configuration, especially in the dynamic mode in pursuit of high transmission performance. To address this shortfall, this contribution develops dynamic OSM/OSSK (D-OSM/OSSK) in the context of MIMO configuration so as to unlock enhanced performance capabilities. Through a combination of rigorous theoretical analysis and simulations, our findings unequivocally demonstrate the superiority of D-OSM/OSSK-MIMO over its counterparts, including OSM/OSSK, spatial modulation (SM), and space shift keying (SSK), while efficiently reducing the hardware cost for user equipment (UE) of consumer electronics.
{"title":"Design of Dynamic Offset Spatial Modulation MIMO for Low-Cost Consumer Electronics Devices","authors":"Yanrui Wang;Yue Xiao;Ming Xiao;Nan Li","doi":"10.1109/TCE.2024.3454178","DOIUrl":"10.1109/TCE.2024.3454178","url":null,"abstract":"A family of offset spatial modulation (OSM) and offset space shift keying (OSSK) techniques has been proposed in (Fang et al., 2019) due to their alleviated requirements for radio frequency (RF) switching, toward an efficient design for low-cost consumer electronic (CE) devices with multiple antennas. Yet, the original structure of OSM/OSSK is based on multiple-input single-output (MISO) design, and so far there is no efficient solution to bridge such system with multiple-input multiple-output (MIMO) configuration, especially in the dynamic mode in pursuit of high transmission performance. To address this shortfall, this contribution develops dynamic OSM/OSSK (D-OSM/OSSK) in the context of MIMO configuration so as to unlock enhanced performance capabilities. Through a combination of rigorous theoretical analysis and simulations, our findings unequivocally demonstrate the superiority of D-OSM/OSSK-MIMO over its counterparts, including OSM/OSSK, spatial modulation (SM), and space shift keying (SSK), while efficiently reducing the hardware cost for user equipment (UE) of consumer electronics.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"70 4","pages":"7526-7534"},"PeriodicalIF":4.3,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142189663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Integrating cutting-edge communication technology with vehicular advancement has led to Vehicular Ad-Hoc Networks (VANETs). VANET architecture facilitates the exchange of vital safety-related messages among vehicles. However, ensuring the authentication and integrity of shared messages over wireless links poses challenges. To resolve the issues, various batch-verifiable authentication schemes have been devised previously. However, existing VANET batch-verifiable authentication schemes utilize number theory-based cryptography, and therefore are vulnerable to quantum computing attacks. Additionally, storing multiple pseudonyms for anonymity incurs storage overhead on vehicles. To address these issues, this paper presents a novel lattice-based dynamic anonymous batch-verifiable authentication scheme. Being a lattice-based design, it is robust against post-quantum threats. To achieve dynamic anonymity, a fuzzy extractor design has been utilized, which removes the storage of multiple pseudonyms. The provable security has been achieved via formal analysis in the random oracle model, and an extensive performance evaluation confirms its efficiency and suitability for VANETs.
{"title":"Dynamic Anonymous Quantum-Secure Batch-Verifiable Authentication Scheme for VANET","authors":"Nahida Majeed Wani;Girraj Kumar Verma;Vinay Chamola","doi":"10.1109/TCE.2024.3453953","DOIUrl":"10.1109/TCE.2024.3453953","url":null,"abstract":"Integrating cutting-edge communication technology with vehicular advancement has led to Vehicular Ad-Hoc Networks (VANETs). VANET architecture facilitates the exchange of vital safety-related messages among vehicles. However, ensuring the authentication and integrity of shared messages over wireless links poses challenges. To resolve the issues, various batch-verifiable authentication schemes have been devised previously. However, existing VANET batch-verifiable authentication schemes utilize number theory-based cryptography, and therefore are vulnerable to quantum computing attacks. Additionally, storing multiple pseudonyms for anonymity incurs storage overhead on vehicles. To address these issues, this paper presents a novel lattice-based dynamic anonymous batch-verifiable authentication scheme. Being a lattice-based design, it is robust against post-quantum threats. To achieve dynamic anonymity, a fuzzy extractor design has been utilized, which removes the storage of multiple pseudonyms. The provable security has been achieved via formal analysis in the random oracle model, and an extensive performance evaluation confirms its efficiency and suitability for VANETs.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"70 4","pages":"7112-7120"},"PeriodicalIF":4.3,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142224475","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-03DOI: 10.1109/TCE.2024.3453890
Shihong Yao;Keyu Pan;Tao Wang;Zhigao Zheng;Jing Jin;Chuli Hu
In large consumer sites, pedestrian re-identification (Re-ID) has the potential to enhance identify loyal consumers and create a more enjoyable shopping experience. Current Re-ID models always rely on some certain pedestrian feature descriptors, including body parts matching and pose key points, to extract part-level features. However, occlusion always causes a tremendous amount of noise and affects the feature representation, thereby significantly degrading the performance of those models. To address this problem, we propose an attribute-aware graph-based network (AAGNet) for Occluded Re-ID. Specifically, we develop a part-attribute feature extractor that maps the manually labeled pedestrian features into word vectors, and combines them with specific body part to obtain both attribute features and part features. The weight information of body parts and attributes are learned through graph convolution networks. Moreover, we introduce an occluded Re-ID dataset called Occluded-Market that can support the subsequent studies of occluded Re-ID. Comparative experimental results evidently demonstrate that the AAGNet shows superior performance in terms of accuracy, efficiency, and robustness on two open-source data sets. Our study can provide data and methodological support for further research on the occluded Re-ID and technological baseline for Re-ID-based commercial analytic applications in large consumer sites. The dataset is available at: github.com/Occluded_Market.
{"title":"AAGNet: Attribute-Aware Graph-Based Network for Occluded Pedestrian Re-Identification","authors":"Shihong Yao;Keyu Pan;Tao Wang;Zhigao Zheng;Jing Jin;Chuli Hu","doi":"10.1109/TCE.2024.3453890","DOIUrl":"10.1109/TCE.2024.3453890","url":null,"abstract":"In large consumer sites, pedestrian re-identification (Re-ID) has the potential to enhance identify loyal consumers and create a more enjoyable shopping experience. Current Re-ID models always rely on some certain pedestrian feature descriptors, including body parts matching and pose key points, to extract part-level features. However, occlusion always causes a tremendous amount of noise and affects the feature representation, thereby significantly degrading the performance of those models. To address this problem, we propose an attribute-aware graph-based network (AAGNet) for Occluded Re-ID. Specifically, we develop a part-attribute feature extractor that maps the manually labeled pedestrian features into word vectors, and combines them with specific body part to obtain both attribute features and part features. The weight information of body parts and attributes are learned through graph convolution networks. Moreover, we introduce an occluded Re-ID dataset called Occluded-Market that can support the subsequent studies of occluded Re-ID. Comparative experimental results evidently demonstrate that the AAGNet shows superior performance in terms of accuracy, efficiency, and robustness on two open-source data sets. Our study can provide data and methodological support for further research on the occluded Re-ID and technological baseline for Re-ID-based commercial analytic applications in large consumer sites. The dataset is available at: github.com/Occluded_Market.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"70 4","pages":"6580-6588"},"PeriodicalIF":4.3,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142224473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-29DOI: 10.1109/TCE.2024.3380085
Prabhat Kumar;Alireza Jolfaei;Krishna Kant
The Tactile Internet (TI) is a logical transition of the Internet, which has progressed from a static, text-based Internet to a multimedia mobile Internet and finally to a Consumer Internet of Things (IoT). The major requirement of any TI applications is low latency, fast transit intervals, high availability, and a high level of security. For instance, latency requirement in Human to Machine (H2M) interactions may vary from < 10 ms up to tens of milliseconds and round-trip latency of 1 ms. This necessitates tactile applications close to end users to minimize delays. Edge Computing (EC) is a resource-rich decentralized platform that offers cloud computing functionalities at cellular base stations near users, saving energy and time on backhaul transmission to cloud servers. In a typical network security architecture of TI, the network administrator establishes network security policies, which segregate network traffic. However, deploying EC at the Internet edge places a strain on network management policies, making them subject to attacks such as Denial-of-Service (DoS), which can harm EC and produce unnecessary network traffic. This type of attack is restricted to EC nodes and has little effect on the backhaul network (such as cloud computing), which is more secure. Therefore, with the growing number of attack vectors, it is essential to develop security solutions for EC to enable computing-based TI applications secure and give application developers more alternatives. The Convergence of Cloud, EC, AI, and blockchain can potentially tackle major shortcomings of TI-driven Consumer IoT, its adoption is still in its infancy, suffering from various issues, such as lack of consensus towards any reference models or best practices.
触觉互联网（TI）是互联网的逻辑过渡，它从静态、基于文本的互联网发展到多媒体移动互联网，最后发展到消费物联网（IoT）。任何 TI 应用的主要要求都是低延迟、快速传输间隔、高可用性和高安全性。例如，人机（H2M）交互的延迟要求可能从 10 毫秒到几十毫秒不等，往返延迟可达 1 毫秒。这就要求触觉应用靠近终端用户，以尽量减少延迟。边缘计算（EC）是一种资源丰富的分散式平台，可在用户附近的蜂窝基站提供云计算功能，从而节省向云服务器回程传输的能源和时间。在典型的 TI 网络安全架构中，网络管理员建立网络安全策略，隔离网络流量。然而，在互联网边缘部署 EC 会给网络管理策略带来压力，使其受到拒绝服务（DoS）等攻击，从而损害 EC 并产生不必要的网络流量。这类攻击仅限于 EC 节点，对更安全的回程网络（如云计算）影响不大。因此，随着攻击载体的日益增多，必须为 EC 开发安全解决方案，以确保基于计算的 TI 应用安全，并为应用开发人员提供更多选择。云、EC、人工智能和区块链的融合有可能解决 TI 驱动的消费类物联网的主要缺点，但其应用仍处于起步阶段，存在各种问题，例如缺乏对任何参考模型或最佳实践的共识。
{"title":"Guest Editorial of the Special Section on Tactile Internet for Consumer Internet of Things Opportunities and Challenges","authors":"Prabhat Kumar;Alireza Jolfaei;Krishna Kant","doi":"10.1109/TCE.2024.3380085","DOIUrl":"https://doi.org/10.1109/TCE.2024.3380085","url":null,"abstract":"The Tactile Internet (TI) is a logical transition of the Internet, which has progressed from a static, text-based Internet to a multimedia mobile Internet and finally to a Consumer Internet of Things (IoT). The major requirement of any TI applications is low latency, fast transit intervals, high availability, and a high level of security. For instance, latency requirement in Human to Machine (H2M) interactions may vary from < 10 ms up to tens of milliseconds and round-trip latency of 1 ms. This necessitates tactile applications close to end users to minimize delays. Edge Computing (EC) is a resource-rich decentralized platform that offers cloud computing functionalities at cellular base stations near users, saving energy and time on backhaul transmission to cloud servers. In a typical network security architecture of TI, the network administrator establishes network security policies, which segregate network traffic. However, deploying EC at the Internet edge places a strain on network management policies, making them subject to attacks such as Denial-of-Service (DoS), which can harm EC and produce unnecessary network traffic. This type of attack is restricted to EC nodes and has little effect on the backhaul network (such as cloud computing), which is more secure. Therefore, with the growing number of attack vectors, it is essential to develop security solutions for EC to enable computing-based TI applications secure and give application developers more alternatives. The Convergence of Cloud, EC, AI, and blockchain can potentially tackle major shortcomings of TI-driven Consumer IoT, its adoption is still in its infancy, suffering from various issues, such as lack of consensus towards any reference models or best practices.","PeriodicalId":13208,"journal":{"name":"IEEE Transactions on Consumer Electronics","volume":"70 2","pages":"4965-4967"},"PeriodicalIF":4.3,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10659259","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142099798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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：