Takaya Miyazawa, Ved P. Kafle, Yusuke Yokota, Yasushi Naruse, Hitoshi Asaeda
The recent rapid advancement of cloud-native networking infrastructure has leveraged the resource virtualization technology of containers to realize diverse microservice-based applications in 5G/6G networks and clouds. Containers drastically enhance the efficiency of computational resource allocation and utilization as compared to the related virtualization technology of Virtual Machines (VMs). The networking environment leveraging both VM and container mixed virtualization technologies makes the most use of them to realize a computational platform whose resources can be dynamically adjusted to a fine granularity. To continuously meet the required levels of quality of services in 5G/6G networks and clouds in that platform, an agile and autonomous data analytics system in the control plane is essential for the accurate prediction of server workloads and dynamic allocation of enough amount of computational resource. In this paper, we introduce a framework, which complies with Recommendation ITU-T Y.3177, for autonomous computational resource control and management. The framework consists of an advanced data analytics system and a resource control system. We propose an architecture for the advanced data analytics system consisting of learning and prediction components. The learning component includes a three-stage intelligent model pipelining with three cascaded machine learning models, nonlinear regression, clustering, and multiple regression. These models determine the fluctuation trends in CPU utilization, classify services with similarities in the trends, and predict the peak CPU utilization of each containerized microservice. We evaluate the proposed models through experiments and numerical analysis. The results prove that the models support agile data analytics, which can complete data processing in the time granularity of seconds and achieve higher prediction accuracy of CPU utilization.
{"title":"Advanced data analytics using three-stage intelligent model pipelining for containerized microservices in 5G networks and beyond","authors":"Takaya Miyazawa, Ved P. Kafle, Yusuke Yokota, Yasushi Naruse, Hitoshi Asaeda","doi":"10.52953/ogkf3616","DOIUrl":"https://doi.org/10.52953/ogkf3616","url":null,"abstract":"The recent rapid advancement of cloud-native networking infrastructure has leveraged the resource virtualization technology of containers to realize diverse microservice-based applications in 5G/6G networks and clouds. Containers drastically enhance the efficiency of computational resource allocation and utilization as compared to the related virtualization technology of Virtual Machines (VMs). The networking environment leveraging both VM and container mixed virtualization technologies makes the most use of them to realize a computational platform whose resources can be dynamically adjusted to a fine granularity. To continuously meet the required levels of quality of services in 5G/6G networks and clouds in that platform, an agile and autonomous data analytics system in the control plane is essential for the accurate prediction of server workloads and dynamic allocation of enough amount of computational resource. In this paper, we introduce a framework, which complies with Recommendation ITU-T Y.3177, for autonomous computational resource control and management. The framework consists of an advanced data analytics system and a resource control system. We propose an architecture for the advanced data analytics system consisting of learning and prediction components. The learning component includes a three-stage intelligent model pipelining with three cascaded machine learning models, nonlinear regression, clustering, and multiple regression. These models determine the fluctuation trends in CPU utilization, classify services with similarities in the trends, and predict the peak CPU utilization of each containerized microservice. We evaluate the proposed models through experiments and numerical analysis. The results prove that the models support agile data analytics, which can complete data processing in the time granularity of seconds and achieve higher prediction accuracy of CPU utilization.","PeriodicalId":274720,"journal":{"name":"ITU Journal on Future and Evolving Technologies","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127311892","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}
Adrian Kliks, Marcin Dryjanski, Vishnu Ram, Leon Wong, Paul Harvey
In this paper we give an overview of an open disaggregated network architecture based on an Open Radio Access Network (O-RAN), including the current work from standards bodies and industry bodies in this area. Based on this architecture, a framework for the automation of xApp development and deployment is proposed. This is then aligned with the key concepts described in ITU-T in terms of the evolution, experimentation, and adaptation of controllers. The various steps in such an aligned workflow, including design, validation, and deployment of xApps, are discussed, and use case examples are provided to illustrate further our position regarding the mechanisms needed to achieve automation.
{"title":"Towards autonomous open radio access networks","authors":"Adrian Kliks, Marcin Dryjanski, Vishnu Ram, Leon Wong, Paul Harvey","doi":"10.52953/gjii3746","DOIUrl":"https://doi.org/10.52953/gjii3746","url":null,"abstract":"In this paper we give an overview of an open disaggregated network architecture based on an Open Radio Access Network (O-RAN), including the current work from standards bodies and industry bodies in this area. Based on this architecture, a framework for the automation of xApp development and deployment is proposed. This is then aligned with the key concepts described in ITU-T in terms of the evolution, experimentation, and adaptation of controllers. The various steps in such an aligned workflow, including design, validation, and deployment of xApps, are discussed, and use case examples are provided to illustrate further our position regarding the mechanisms needed to achieve automation.","PeriodicalId":274720,"journal":{"name":"ITU Journal on Future and Evolving Technologies","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127653700","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}
Cloud and edge computing, distributed AI, and most recently 5G/6G communications are coming together and changing the way we collaborate, connect and interact. A new generation of AI-powered robots are also expected to be facilitated by these digital technological breakthroughs. Robots are supposed to tackle unknown situations and adapt in the long term by collaborating, connecting and interacting with the digital world. Such applications generate versatile, perpetuated and rapidly changing transmission demands to the network. Traditional network resource management is insufficient in supporting such traffic to meet the QoS. In this paper, we go a step further, in addition to the effort on the network side for traffic engineering; we also work on the application side to shape the traffic within non-public networks. We present an initial development for the proposed intent-based deployment for robotic applications.
{"title":"Intent-based deployment for robot applications in 5G-enabled non-public networks","authors":"Renxi Qiu, Dayou Li, Adri�n Lend�nez Ib��ez, Zhao Xu, Rafa L�pez Taraz�n","doi":"10.52953/aymi1991","DOIUrl":"https://doi.org/10.52953/aymi1991","url":null,"abstract":"Cloud and edge computing, distributed AI, and most recently 5G/6G communications are coming together and changing the way we collaborate, connect and interact. A new generation of AI-powered robots are also expected to be facilitated by these digital technological breakthroughs. Robots are supposed to tackle unknown situations and adapt in the long term by collaborating, connecting and interacting with the digital world. Such applications generate versatile, perpetuated and rapidly changing transmission demands to the network. Traditional network resource management is insufficient in supporting such traffic to meet the QoS. In this paper, we go a step further, in addition to the effort on the network side for traffic engineering; we also work on the application side to shape the traffic within non-public networks. We present an initial development for the proposed intent-based deployment for robotic applications.","PeriodicalId":274720,"journal":{"name":"ITU Journal on Future and Evolving Technologies","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128337007","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}
Recently, Terahertz (THz) band communications at various atmospheric altitudes have been studied due to larger bandwidth availability and reduced water vapor concentrations at higher atmospheric altitudes as compared to sea level. In this paper, as special cases of 6G aerial communication networks, we consider: (1) Low Altitude Platform-to-High Altitude Platform (LAP-to-HAP), (2) HAP-to-HAP, and (3) HAP-to-Satellite (HAP-to-SAT) GHz-to-THz broadband communications over (1-1000) GHz by analyzing total path loss and total usable bandwidth. For obtaining realistic absorption loss at practical altitudes, we employ the International Telecommunications Union's (ITU) model using the standard weather profile. We consider four practical carrier frequencies offering low absorption loss values i.e., f1 = 0.140 THz (D band: 110-170 GHz), f2 = 0.300 THz (275-325 GHz), f3 = 0.750 THz, and f4 = 0.875 THz for analyzing total path loss. Numerical results show that due to improved atmospheric conditions particularly above 23 km altitudes, increasing the Rx-HAP altitude in HAP-to-HAP communications promises a lower total path loss of up to 7.7 %, even at the cost of an increase in the Tx-Rx-HAP distance from 1 km to 34 km, promising Tbps rates for 6G non-terrestrial communications. Additionally, total usable bandwidth analysis demonstrates that with total antenna gains of 80 dBi, bandwidth in the order of 100s of GHz is usable for the LAP-to-HAP scenario, the entire considered broadband is usable for the HAP-to-HAP scenario between 16 km to 50 km, and the HAP-to-SAT scenario between a HAP at 19 km and SAT at 100 km, truly showcasing the potential of employing GHz-to-THz broadband communications cognitively over (1-1000) GHz for various practical 6G non-terrestrial networks.
{"title":"GHz-to-THz broadband communications for 6G non-terrestrial networks","authors":"Akhtar Saeed, Hilal Esra Yaldiz, Fatih Alagoz","doi":"10.52953/aoky1032","DOIUrl":"https://doi.org/10.52953/aoky1032","url":null,"abstract":"Recently, Terahertz (THz) band communications at various atmospheric altitudes have been studied due to larger bandwidth availability and reduced water vapor concentrations at higher atmospheric altitudes as compared to sea level. In this paper, as special cases of 6G aerial communication networks, we consider: (1) Low Altitude Platform-to-High Altitude Platform (LAP-to-HAP), (2) HAP-to-HAP, and (3) HAP-to-Satellite (HAP-to-SAT) GHz-to-THz broadband communications over (1-1000) GHz by analyzing total path loss and total usable bandwidth. For obtaining realistic absorption loss at practical altitudes, we employ the International Telecommunications Union's (ITU) model using the standard weather profile. We consider four practical carrier frequencies offering low absorption loss values i.e., f1 = 0.140 THz (D band: 110-170 GHz), f2 = 0.300 THz (275-325 GHz), f3 = 0.750 THz, and f4 = 0.875 THz for analyzing total path loss. Numerical results show that due to improved atmospheric conditions particularly above 23 km altitudes, increasing the Rx-HAP altitude in HAP-to-HAP communications promises a lower total path loss of up to 7.7 %, even at the cost of an increase in the Tx-Rx-HAP distance from 1 km to 34 km, promising Tbps rates for 6G non-terrestrial communications. Additionally, total usable bandwidth analysis demonstrates that with total antenna gains of 80 dBi, bandwidth in the order of 100s of GHz is usable for the LAP-to-HAP scenario, the entire considered broadband is usable for the HAP-to-HAP scenario between 16 km to 50 km, and the HAP-to-SAT scenario between a HAP at 19 km and SAT at 100 km, truly showcasing the potential of employing GHz-to-THz broadband communications cognitively over (1-1000) GHz for various practical 6G non-terrestrial networks.","PeriodicalId":274720,"journal":{"name":"ITU Journal on Future and Evolving Technologies","volume":"164 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121915858","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}
Wireless technology is expected to become a fundamental enabler to improve the efficiency, safety, and revenues of advanced manufacturing processes, as well as to realize new paradigms such as digital twins. The extremely challenging industrial scenario requires some technological shifts such as the adoption of the so far unexplored THz band. The purpose of this paper is to provide an overview of THz networks applied to the Industrial Internet of Things (IIoT). First, the main requirements of future industrial THz-based networks, challenges, and state-of-the-art are described. Subsequently, the key enabling technologies are introduced and discussed. Finally, we present some research directions for THz-based industrial networks.
{"title":"Terahertz networks for future Industrial Internet of Things","authors":"Sara Cavallero, Nicol� Decarli, Giampaolo Cuozzo, Chiara Buratti, Davide Dardari, Roberto Verdone","doi":"10.52953/athi4610","DOIUrl":"https://doi.org/10.52953/athi4610","url":null,"abstract":"Wireless technology is expected to become a fundamental enabler to improve the efficiency, safety, and revenues of advanced manufacturing processes, as well as to realize new paradigms such as digital twins. The extremely challenging industrial scenario requires some technological shifts such as the adoption of the so far unexplored THz band. The purpose of this paper is to provide an overview of THz networks applied to the Industrial Internet of Things (IIoT). First, the main requirements of future industrial THz-based networks, challenges, and state-of-the-art are described. Subsequently, the key enabling technologies are introduced and discussed. Finally, we present some research directions for THz-based industrial networks.","PeriodicalId":274720,"journal":{"name":"ITU Journal on Future and Evolving Technologies","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134395713","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}
Gianfranco Ciccarella, Romeo Giuliano, Franco Mazzenga, Francesco Vatalaro, Alessandro Vizzarri
This paper analyzes some of the telecommunication companies' (telcos) domestic networks' main objectives and issues related to the application services Quality of Experience (QoE), i.e., "technical QoE," and the economic sustainability of the Very High Capacity (VHC) networks. Telco domestic networks and Over The Top (OTT) networks are the Internet segments connecting the end-user equipment to the servers/clouds that provide application services. With the advent of VHC networks, telcos afford difficulties in managing effectively and efficiently the application services and their business sustainability. This paper affords some issues related to traditional telcos' domestic network architectures and the approach to managing the application services' performance improvement. Some telcos started changing their architectures, although a massive transformation had not yet started in the industry. Traditionally, telcos focus on network services, i.e., the transport of IP packets. Performance improvement is obtained by QoS-based traffic management techniques, such as bandwidth reservation and packet prioritization. To manage application performance improvement, the use of layer 4 techniques and Edge Cloud Computing (ECC) is effective, as demonstrated by the multiyear experience of OTTs which already use these technologies. In telco domestic networks, layer 2 tunnels increase the complexity of deploying ECC. However, some vendors provide non-standard solutions to make the IP layer 3 user plane visible and to deploy ECC. ECC is the key factor for the transformation. It is a mini/micro-data center that distributes some applications and content closer to the end users. The distribution can provide a paradigm shift in a telco's business. In the paper, we first highlight the need for regulators to appreciate the need to encourage this industry transformation fully.
本文分析了一些电信公司(telcos)国内网络的主要目标和与应用服务体验质量(QoE)(即“技术QoE”)和极高容量(VHC)网络的经济可持续性相关的问题。电信国内网络和OTT (Over The Top)网络是将最终用户设备连接到提供应用服务的服务器/云的互联网段。随着VHC网络的出现，电信公司在有效和高效地管理应用服务及其业务可持续性方面遇到了困难。本文提出了国内传统电信运营商网络架构中存在的一些问题，以及应用服务性能提升的管理方法。一些电信公司开始改变他们的架构，尽管行业还没有开始大规模的转型。传统上，电信公司专注于网络服务，即IP数据包的传输。性能改进是通过基于qos的流量管理技术，如带宽预留和数据包优先级来实现的。为了管理应用程序性能改进，使用第4层技术和边缘云计算(ECC)是有效的，正如已经使用这些技术的ott多年的经验所证明的那样。在电信国内网络中，二层隧道增加了部署ECC的复杂性。然而，一些供应商提供了非标准的解决方案，使IP三层用户平面可见，并部署ECC。ECC是实现这一转变的关键因素。它是一个迷你/微型数据中心，将一些应用程序和内容分发到更接近最终用户的地方。分销可以为电信公司的业务提供范式转变。在本文中，我们首先强调监管机构需要充分认识到鼓励这一行业转型的必要性。
{"title":"Why and how edge cloud computing can address performance and economic sustainability issues for telco domestic networks","authors":"Gianfranco Ciccarella, Romeo Giuliano, Franco Mazzenga, Francesco Vatalaro, Alessandro Vizzarri","doi":"10.52953/omkr7642","DOIUrl":"https://doi.org/10.52953/omkr7642","url":null,"abstract":"This paper analyzes some of the telecommunication companies' (telcos) domestic networks' main objectives and issues related to the application services Quality of Experience (QoE), i.e., \"technical QoE,\" and the economic sustainability of the Very High Capacity (VHC) networks. Telco domestic networks and Over The Top (OTT) networks are the Internet segments connecting the end-user equipment to the servers/clouds that provide application services. With the advent of VHC networks, telcos afford difficulties in managing effectively and efficiently the application services and their business sustainability. This paper affords some issues related to traditional telcos' domestic network architectures and the approach to managing the application services' performance improvement. Some telcos started changing their architectures, although a massive transformation had not yet started in the industry. Traditionally, telcos focus on network services, i.e., the transport of IP packets. Performance improvement is obtained by QoS-based traffic management techniques, such as bandwidth reservation and packet prioritization. To manage application performance improvement, the use of layer 4 techniques and Edge Cloud Computing (ECC) is effective, as demonstrated by the multiyear experience of OTTs which already use these technologies. In telco domestic networks, layer 2 tunnels increase the complexity of deploying ECC. However, some vendors provide non-standard solutions to make the IP layer 3 user plane visible and to deploy ECC. ECC is the key factor for the transformation. It is a mini/micro-data center that distributes some applications and content closer to the end users. The distribution can provide a paradigm shift in a telco's business. In the paper, we first highlight the need for regulators to appreciate the need to encourage this industry transformation fully.","PeriodicalId":274720,"journal":{"name":"ITU Journal on Future and Evolving Technologies","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115118550","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}
When it comes to the security of the Internet of Things (IoT), securing their communications is paramount. In multi-hop networks, nodes relay information amongst themselves, opening the data up to tampering by an intermediate device. To detect and avoid such malicious entities, we grant nodes the ability to analyse their neighbours behaviour. Through the use of consensus-based validation, based upon the blockchain's miners, all nodes can agree on the trustworthiness of all devices in the network. By expressing this through a node's reputation, it is possible to identify malicious devices and isolate them from network activities. By incorporating this metric into a multi-hop routing protocol such as AODV, we can influence the path selection process. Instead of defining the best route based upon overall length, we can choose the most reputable path available, thus traversing trustworthy devices. By performing extensive analyses through multiple simulated scenarios, we can identify a decrease in packet drop rates compared to AODV by approximately 48% and 38% when subjected to black hole attacks with 30 and 100 node networks respectively. Furthermore, by subjecting our system to varying degrees of grey holes, we can confirm its adaptability to different types of threats.
{"title":"A consensus-based approach to reputational routing in multi-hop networks","authors":"Edward Staddon, Valeria Loscri, Nathalie Mitton","doi":"10.52953/ixzf4584","DOIUrl":"https://doi.org/10.52953/ixzf4584","url":null,"abstract":"When it comes to the security of the Internet of Things (IoT), securing their communications is paramount. In multi-hop networks, nodes relay information amongst themselves, opening the data up to tampering by an intermediate device. To detect and avoid such malicious entities, we grant nodes the ability to analyse their neighbours behaviour. Through the use of consensus-based validation, based upon the blockchain's miners, all nodes can agree on the trustworthiness of all devices in the network. By expressing this through a node's reputation, it is possible to identify malicious devices and isolate them from network activities. By incorporating this metric into a multi-hop routing protocol such as AODV, we can influence the path selection process. Instead of defining the best route based upon overall length, we can choose the most reputable path available, thus traversing trustworthy devices. By performing extensive analyses through multiple simulated scenarios, we can identify a decrease in packet drop rates compared to AODV by approximately 48% and 38% when subjected to black hole attacks with 30 and 100 node networks respectively. Furthermore, by subjecting our system to varying degrees of grey holes, we can confirm its adaptability to different types of threats.","PeriodicalId":274720,"journal":{"name":"ITU Journal on Future and Evolving Technologies","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116369398","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 5G and beyond on the horizon, ultra-fast and low latency data transmission on the cloud and via the Internet will enable more intelligent and interactive medical and health-care applications. This paper presents a review of 5G technologies and their related applications in the health-care sector. The introduction to 5G technology includes software defined network, 5G architecture and edge computing. The second part of the paper then presents the opportunities provided by 5G to the health-care sector and employs medical imaging applications as central examples to demonstrate the impacts of 5G and the cloud. Finally, this paper summarize the benefits brought by 5G and cloud computing to the health-care sector.
{"title":"Boosting smarter digital health care with 5G and beyond networks","authors":"Enjie Liu, Youbing Zhao, Abimbola Efunogbon","doi":"10.52953/gjnn6958","DOIUrl":"https://doi.org/10.52953/gjnn6958","url":null,"abstract":"With 5G and beyond on the horizon, ultra-fast and low latency data transmission on the cloud and via the Internet will enable more intelligent and interactive medical and health-care applications. This paper presents a review of 5G technologies and their related applications in the health-care sector. The introduction to 5G technology includes software defined network, 5G architecture and edge computing. The second part of the paper then presents the opportunities provided by 5G to the health-care sector and employs medical imaging applications as central examples to demonstrate the impacts of 5G and the cloud. Finally, this paper summarize the benefits brought by 5G and cloud computing to the health-care sector.","PeriodicalId":274720,"journal":{"name":"ITU Journal on Future and Evolving Technologies","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126039609","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}
Xi Lan, Jalil Taghia, Farnaz Moradi, M. Khoshkholghi, Edvin Listo Zec, Olof Mogren, Toktam Mahmoodi, Andreas Johnsson
Telecom vendors and operators deliver services with strict requirements on performance, over complex and sometimes partly shared network infrastructures. A key enabler for network and service management in such environments is knowledge sharing, and the use of data-driven models for performance prediction, forecasting, and troubleshooting. In this paper, we outline a multi-operator service metrics prediction framework using federated learning that allows privacy-preserved knowledge-sharing across operators for improved model performance, and also reduced requirements on data transfer within an operator network. Federated learning is compared against local and central learning strategies for multi-operator performance prediction, and it is shown to balance the requirements on data privacy, model performance, and the network overhead. Further, the paper provides insights on how data heterogeneity affects model performance, where the conclusion is that standard federated learning has certain robustness to data heterogeneity. Finally, we discuss the challenges related to training a federated learning model with a limited budget on the communication rounds. The evaluation is performed using a set of realistic publicly available data traces, that are adapted specifically for the purpose of studying multi-operator service performance prediction.
{"title":"Federated learning for performance prediction in multi-operator environments","authors":"Xi Lan, Jalil Taghia, Farnaz Moradi, M. Khoshkholghi, Edvin Listo Zec, Olof Mogren, Toktam Mahmoodi, Andreas Johnsson","doi":"10.52953/pfyz9165","DOIUrl":"https://doi.org/10.52953/pfyz9165","url":null,"abstract":"Telecom vendors and operators deliver services with strict requirements on performance, over complex and sometimes partly shared network infrastructures. A key enabler for network and service management in such environments is knowledge sharing, and the use of data-driven models for performance prediction, forecasting, and troubleshooting. In this paper, we outline a multi-operator service metrics prediction framework using federated learning that allows privacy-preserved knowledge-sharing across operators for improved model performance, and also reduced requirements on data transfer within an operator network. Federated learning is compared against local and central learning strategies for multi-operator performance prediction, and it is shown to balance the requirements on data privacy, model performance, and the network overhead. Further, the paper provides insights on how data heterogeneity affects model performance, where the conclusion is that standard federated learning has certain robustness to data heterogeneity. Finally, we discuss the challenges related to training a federated learning model with a limited budget on the communication rounds. The evaluation is performed using a set of realistic publicly available data traces, that are adapted specifically for the purpose of studying multi-operator service performance prediction.","PeriodicalId":274720,"journal":{"name":"ITU Journal on Future and Evolving Technologies","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130405156","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}
Improving transportation efficiency and on-road safety using Intelligent Transportation Systems (ITSs) has become crucial as road congestion and vehicle complexity increase coupled with ongoing rapid development and deployment of electric vehicles across the globe. Recent advances in computer systems and wireless communications have ushered in more possibilities for smart solutions to road traffic safety, congestion reduction, convenience, and overall efficiency. The evolution and deployment of 5G have opened up new technologies and features that can provide the much needed high-mobility wireless networks for the emerging Internet of Vehicles (IoV). The application of AI consisting of Deep Learning (DL), Machine Learning (ML) and Swarm Intelligence (SI) techniques have emerged in both conventional and vehicular wireless networks with strong promises towards enhancing traditional data-centric methods. Particularly, in the application domains of IoV, big data is frequently generated from various sources within the vehicular communication environment. The collected big data is usually processed and used for both safety and infotainment services including routing, broadening drivers' awareness, traffic mobility prediction for hazardous situation avoidance to improve overall safety and passenger comfort, and general quality of road experience. Applying data-driven methods enables AI to address high mobility and dynamic vehicular communications and network issues facing traditional solutions and approaches like network optimization techniques and conventional control loop design. This study provides a concise review of DL, ML and SI techniques and applications that are currently being explored by different research efforts within the application area of vehicular networks. The paper further discusses the strengths and weaknesses of the proposed AI-based solutions for the IoV networks.
{"title":"Artificial intelligence support for 5G/6G-enabled Internet of Vehicles networks: An overview","authors":"Elias Eze, Joy Eze","doi":"10.52953/iezn8770","DOIUrl":"https://doi.org/10.52953/iezn8770","url":null,"abstract":"Improving transportation efficiency and on-road safety using Intelligent Transportation Systems (ITSs) has become crucial as road congestion and vehicle complexity increase coupled with ongoing rapid development and deployment of electric vehicles across the globe. Recent advances in computer systems and wireless communications have ushered in more possibilities for smart solutions to road traffic safety, congestion reduction, convenience, and overall efficiency. The evolution and deployment of 5G have opened up new technologies and features that can provide the much needed high-mobility wireless networks for the emerging Internet of Vehicles (IoV). The application of AI consisting of Deep Learning (DL), Machine Learning (ML) and Swarm Intelligence (SI) techniques have emerged in both conventional and vehicular wireless networks with strong promises towards enhancing traditional data-centric methods. Particularly, in the application domains of IoV, big data is frequently generated from various sources within the vehicular communication environment. The collected big data is usually processed and used for both safety and infotainment services including routing, broadening drivers' awareness, traffic mobility prediction for hazardous situation avoidance to improve overall safety and passenger comfort, and general quality of road experience. Applying data-driven methods enables AI to address high mobility and dynamic vehicular communications and network issues facing traditional solutions and approaches like network optimization techniques and conventional control loop design. This study provides a concise review of DL, ML and SI techniques and applications that are currently being explored by different research efforts within the application area of vehicular networks. The paper further discusses the strengths and weaknesses of the proposed AI-based solutions for the IoV networks.","PeriodicalId":274720,"journal":{"name":"ITU Journal on Future and Evolving Technologies","volume":"165 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120840340","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：