Pub Date : 2021-09-20DOI: 10.1109/BalkanCom53780.2021.9593121
N. Gligoric, S. Krco, D. Drajić
The paper presents industry 4.0 scenario for predictive maintenance based on 3-axis accelerometer sensors and AI platform for monitoring of the machines. The prototype of the service is developed for a company to monitor their machinery for manufacturing of enameled, stainless steel and non-stick cookware (using vibration sensors and data collected from processes). The goal was to provide proof of concept of the predictive maintenance in an industry environment. To achieve this, we have created a mean for evaluating longevity of the equipment using traveled distance as a measure. The travel distance represents a level of vibration that sensor detects while listening to a machine operations in a long term.
{"title":"Digital transformation in Industry 4.0 Using Vibration Sensors and Machine Learning","authors":"N. Gligoric, S. Krco, D. Drajić","doi":"10.1109/BalkanCom53780.2021.9593121","DOIUrl":"https://doi.org/10.1109/BalkanCom53780.2021.9593121","url":null,"abstract":"The paper presents industry 4.0 scenario for predictive maintenance based on 3-axis accelerometer sensors and AI platform for monitoring of the machines. The prototype of the service is developed for a company to monitor their machinery for manufacturing of enameled, stainless steel and non-stick cookware (using vibration sensors and data collected from processes). The goal was to provide proof of concept of the predictive maintenance in an industry environment. To achieve this, we have created a mean for evaluating longevity of the equipment using traveled distance as a measure. The travel distance represents a level of vibration that sensor detects while listening to a machine operations in a long term.","PeriodicalId":115090,"journal":{"name":"2021 International Balkan Conference on Communications and Networking (BalkanCom)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130367637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-20DOI: 10.1109/BalkanCom53780.2021.9593260
Fatih Yürekli, Tolga Girici
Due to their flexible deployment and line-of-sight channel conditions, drone base stations is a promising technology. Recent developments in Massive MIMO tranmission with millimeter-wave beamforming also enables high data rates and enables simultaneous transmission to multiple ground users. In this work we consider the problem of deployment (i.e. positioning) of drone base stations and analog MIMO beamforming that maximizes SINR. Simulation results reveal that careful K-Means clustering of ground users, together with DFT-based beamforming provides a promising performance.
{"title":"Optimal Precoding and Deployment of Millimeter-Wave Drone Base Stations","authors":"Fatih Yürekli, Tolga Girici","doi":"10.1109/BalkanCom53780.2021.9593260","DOIUrl":"https://doi.org/10.1109/BalkanCom53780.2021.9593260","url":null,"abstract":"Due to their flexible deployment and line-of-sight channel conditions, drone base stations is a promising technology. Recent developments in Massive MIMO tranmission with millimeter-wave beamforming also enables high data rates and enables simultaneous transmission to multiple ground users. In this work we consider the problem of deployment (i.e. positioning) of drone base stations and analog MIMO beamforming that maximizes SINR. Simulation results reveal that careful K-Means clustering of ground users, together with DFT-based beamforming provides a promising performance.","PeriodicalId":115090,"journal":{"name":"2021 International Balkan Conference on Communications and Networking (BalkanCom)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114252368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-20DOI: 10.1109/BalkanCom53780.2021.9593239
Lazar Berbakov, Nikola M. Tomasevic
Modern manufacturing and process industry require new adaptive and flexible production system in order to cope with the production requirements of the 21st century. In particular, fast setup of production lines, improved product quality and production efficiency are seen as some of the benefits of adopting Industry 4.0 approaches. In order to achieve the aforementioned goals, there is a need to intelligently process the data collected by using IIoT system and provide results that can be used to act upon the equipment, industrial processes and end products. In this paper, the aim is to present a preliminary architecture which will be used to support the goals of H2020 EU project AI-PROFICIENT. The architecture is divided into multiple layers: system edge, cloud platform and human interaction which comprise functionalities related to data integration, processing and visualization.
{"title":"Internet of Things Platform Architecture for Smart Factories","authors":"Lazar Berbakov, Nikola M. Tomasevic","doi":"10.1109/BalkanCom53780.2021.9593239","DOIUrl":"https://doi.org/10.1109/BalkanCom53780.2021.9593239","url":null,"abstract":"Modern manufacturing and process industry require new adaptive and flexible production system in order to cope with the production requirements of the 21st century. In particular, fast setup of production lines, improved product quality and production efficiency are seen as some of the benefits of adopting Industry 4.0 approaches. In order to achieve the aforementioned goals, there is a need to intelligently process the data collected by using IIoT system and provide results that can be used to act upon the equipment, industrial processes and end products. In this paper, the aim is to present a preliminary architecture which will be used to support the goals of H2020 EU project AI-PROFICIENT. The architecture is divided into multiple layers: system edge, cloud platform and human interaction which comprise functionalities related to data integration, processing and visualization.","PeriodicalId":115090,"journal":{"name":"2021 International Balkan Conference on Communications and Networking (BalkanCom)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123989744","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-20DOI: 10.1109/balkancom53780.2021.9593159
{"title":"[Copyright notice]","authors":"","doi":"10.1109/balkancom53780.2021.9593159","DOIUrl":"https://doi.org/10.1109/balkancom53780.2021.9593159","url":null,"abstract":"","PeriodicalId":115090,"journal":{"name":"2021 International Balkan Conference on Communications and Networking (BalkanCom)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124207821","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-20DOI: 10.1109/BalkanCom53780.2021.9593127
Christodoulos Skouroumounis, I. Krikidis
User mobility in cellular networks deployed in finite areas, results in the non-uniform spatial distribution of the mobile users (MUs), necessitating the effective deployment of base stations (BSs) in order to enhance the network connectivity. In this paper, we analyze the optimal non-uniform distribution of the BSs that provides ultra-reliable connectivity in the context of finite-area cellular networks, where the MUs employ the random waypoint (RWP) model. By using stochastic geometry tools, we first establish an analytical and tractable mathematical framework to investigate the uplink success probability in the context of non-uniform cellular networks. Furthermore, analytical expressions for the moments of the conditional success probability are derived, and a simple approximation of the meta-distribution (MD) is calculated, leveraging the moment-matching method with the Beta-distribution. Finally, we exploit the MD as a criterion for defining the optimal BS distribution, conditioned on the user’s mobility. Our results reveal that the developed framework provides guidance in the design of cellular networks to efficiently determine the optimal network spatial deployment that offer ultra-reliable connectivity based on the mobility of MUs.
{"title":"Meta Distribution-optimal Base Station Deployment for Finite-Area Mobile Networks","authors":"Christodoulos Skouroumounis, I. Krikidis","doi":"10.1109/BalkanCom53780.2021.9593127","DOIUrl":"https://doi.org/10.1109/BalkanCom53780.2021.9593127","url":null,"abstract":"User mobility in cellular networks deployed in finite areas, results in the non-uniform spatial distribution of the mobile users (MUs), necessitating the effective deployment of base stations (BSs) in order to enhance the network connectivity. In this paper, we analyze the optimal non-uniform distribution of the BSs that provides ultra-reliable connectivity in the context of finite-area cellular networks, where the MUs employ the random waypoint (RWP) model. By using stochastic geometry tools, we first establish an analytical and tractable mathematical framework to investigate the uplink success probability in the context of non-uniform cellular networks. Furthermore, analytical expressions for the moments of the conditional success probability are derived, and a simple approximation of the meta-distribution (MD) is calculated, leveraging the moment-matching method with the Beta-distribution. Finally, we exploit the MD as a criterion for defining the optimal BS distribution, conditioned on the user’s mobility. Our results reveal that the developed framework provides guidance in the design of cellular networks to efficiently determine the optimal network spatial deployment that offer ultra-reliable connectivity based on the mobility of MUs.","PeriodicalId":115090,"journal":{"name":"2021 International Balkan Conference on Communications and Networking (BalkanCom)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128510807","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-20DOI: 10.1109/balkancom53780.2021.9593189
{"title":"Conference Organization Committee Members","authors":"","doi":"10.1109/balkancom53780.2021.9593189","DOIUrl":"https://doi.org/10.1109/balkancom53780.2021.9593189","url":null,"abstract":"","PeriodicalId":115090,"journal":{"name":"2021 International Balkan Conference on Communications and Networking (BalkanCom)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125036299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-20DOI: 10.1109/BalkanCom53780.2021.9593218
A. Zervopoulos, K. Oikonomou
The advent of Software-Defined Networking (SDN) brings about a major shift in network management but also introduces various challenges, like scalability, giving rise to physically distributed approaches towards a logically centralized controller. When deploying multiple controllers, where to place them in the network in order to minimize response times and overhead is an issue that naturally arises. This has been termed as the Controller Placement Problem (CPP). The CPP can be viewed as an extension of the k-median problem, if the number of controllers is known. However, unlike the k-median problem, assigning switches to the available controllers is a challenging task, as even the reassignment of a single switch can affect the performance of others. In this work, the CPP is studied in the context of tree topologies, where it is shown that reassigning a switch only affects its directly adjacent switches. Additionally, a migration-based approach for solving the CPP is considered, consisting of two parts: (i) the controller migration policy; and (ii) the switch migration policy. The controller migration policy is based on an existing service migration approach, whereas two different switch migration policies are proposed and studied in this paper. Simulation results demonstrate that in tree topologies, switch migrations offer limited benefit. This is attributed to the limited opportunities for switch migration that arise because of the topological aspects of trees. On the other hand, the controller migration policy suffices to closely approximate optimal solutions.
{"title":"A Migration-Based Approach for the SDN Controller Placement Problem in Tree Topologies","authors":"A. Zervopoulos, K. Oikonomou","doi":"10.1109/BalkanCom53780.2021.9593218","DOIUrl":"https://doi.org/10.1109/BalkanCom53780.2021.9593218","url":null,"abstract":"The advent of Software-Defined Networking (SDN) brings about a major shift in network management but also introduces various challenges, like scalability, giving rise to physically distributed approaches towards a logically centralized controller. When deploying multiple controllers, where to place them in the network in order to minimize response times and overhead is an issue that naturally arises. This has been termed as the Controller Placement Problem (CPP). The CPP can be viewed as an extension of the k-median problem, if the number of controllers is known. However, unlike the k-median problem, assigning switches to the available controllers is a challenging task, as even the reassignment of a single switch can affect the performance of others. In this work, the CPP is studied in the context of tree topologies, where it is shown that reassigning a switch only affects its directly adjacent switches. Additionally, a migration-based approach for solving the CPP is considered, consisting of two parts: (i) the controller migration policy; and (ii) the switch migration policy. The controller migration policy is based on an existing service migration approach, whereas two different switch migration policies are proposed and studied in this paper. Simulation results demonstrate that in tree topologies, switch migrations offer limited benefit. This is attributed to the limited opportunities for switch migration that arise because of the topological aspects of trees. On the other hand, the controller migration policy suffices to closely approximate optimal solutions.","PeriodicalId":115090,"journal":{"name":"2021 International Balkan Conference on Communications and Networking (BalkanCom)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131610448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-20DOI: 10.1109/BalkanCom53780.2021.9593210
M. Haghshenas, Mattia D’Adda, Francesco Linsalata, Luca Barbieri, M. Nicoli, M. Magarini
It is proved that in a rich scattering environment, where the channel can be approximated as a zero-mean circularly symmetric complex Gaussian, Zero-Forcing Beamforming (ZFBF) performs as a sub-optimal spatial multiplexing solution that asymptotically approaches the optimal capacity. In fact, it may benefit from multi-user diversity and spatial multiplexing to increase the sum-rate capacity. However, its efficiency at millimeter wave (mmW), where the propagation suffers from high path loss and blockage, is still an open issue. In this work, we analyze ZFBF performance in a real Infrastructure-to-Vehicles (I2V) scenario at mmW. Our numerical results show that the environment’s geometry significantly affects the propagation pattern and thereby reduces the efficiency of ZFBF in terms of multiplexing. On the other hand, it still takes advantage of multi-user diversity to improve capacity.
{"title":"On the Performance of Zero-forcing Beamforming in a Real I2V Scenario at Millimiter Wave","authors":"M. Haghshenas, Mattia D’Adda, Francesco Linsalata, Luca Barbieri, M. Nicoli, M. Magarini","doi":"10.1109/BalkanCom53780.2021.9593210","DOIUrl":"https://doi.org/10.1109/BalkanCom53780.2021.9593210","url":null,"abstract":"It is proved that in a rich scattering environment, where the channel can be approximated as a zero-mean circularly symmetric complex Gaussian, Zero-Forcing Beamforming (ZFBF) performs as a sub-optimal spatial multiplexing solution that asymptotically approaches the optimal capacity. In fact, it may benefit from multi-user diversity and spatial multiplexing to increase the sum-rate capacity. However, its efficiency at millimeter wave (mmW), where the propagation suffers from high path loss and blockage, is still an open issue. In this work, we analyze ZFBF performance in a real Infrastructure-to-Vehicles (I2V) scenario at mmW. Our numerical results show that the environment’s geometry significantly affects the propagation pattern and thereby reduces the efficiency of ZFBF in terms of multiplexing. On the other hand, it still takes advantage of multi-user diversity to improve capacity.","PeriodicalId":115090,"journal":{"name":"2021 International Balkan Conference on Communications and Networking (BalkanCom)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129471598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-20DOI: 10.1109/BalkanCom53780.2021.9593224
Dandan Li, Jiangfeng Li, Xinhua Zeng, V. Stanković, L. Stanković, Qingjiang Shi
The rapidly expansion of Internet of Things (IoT) has ignited renewed interest in energy disaggregation via non-intrusive load monitoring (NILM). Compared to the more frequent NILM approach of training one model for each appliance, this paper proposes a multi-label learning approach based on the widely cited sequence2point convolutional neural network (CNN). Using the smart meter readings collected in an office building, we demonstrate the accuracy and practicality of the proposed network compared to start-of-the-art one-to-one NILM models.
{"title":"Non-Intrusive Load Monitoring for Multi-objects in Smart Building","authors":"Dandan Li, Jiangfeng Li, Xinhua Zeng, V. Stanković, L. Stanković, Qingjiang Shi","doi":"10.1109/BalkanCom53780.2021.9593224","DOIUrl":"https://doi.org/10.1109/BalkanCom53780.2021.9593224","url":null,"abstract":"The rapidly expansion of Internet of Things (IoT) has ignited renewed interest in energy disaggregation via non-intrusive load monitoring (NILM). Compared to the more frequent NILM approach of training one model for each appliance, this paper proposes a multi-label learning approach based on the widely cited sequence2point convolutional neural network (CNN). Using the smart meter readings collected in an office building, we demonstrate the accuracy and practicality of the proposed network compared to start-of-the-art one-to-one NILM models.","PeriodicalId":115090,"journal":{"name":"2021 International Balkan Conference on Communications and Networking (BalkanCom)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126087878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-20DOI: 10.1109/BalkanCom53780.2021.9593257
G. Ropokis
We focus on the problem of optimal power allocation for a two user interference channel characterized by mixed Channel State Information (CSI), which includes instantaneous information for the direct channels and statistical information for the interference channels. For this model, we introduce a general framework for optimizing the power allocation such as to maximize some generic Quality of Service (QoS) performance metric (or equivalently minimize some cost function). We model this problem as a function approximation problem where the function to be learned is the mapping between CSI and the solution to the optimization problem. We then tackle this problem borrowing ideas from ensemble learning. In particular, using generalized linear models (which are characterized by low complexity and can be implemented even at network nodes characterized by strong computational limitations), we produce different weak learners for learning to solve the considered optimization problem and based on ensemble learning theory, we combine such learners to produce stronger learners. We assess the performance of our framework by applying it on a particular resource allocation problem, and the obtained performance results indicate that the proposed approach can deliver near-optimal performance.
{"title":"An ensemble learning framework for distributed resource allocation in inteference channels: The two user case","authors":"G. Ropokis","doi":"10.1109/BalkanCom53780.2021.9593257","DOIUrl":"https://doi.org/10.1109/BalkanCom53780.2021.9593257","url":null,"abstract":"We focus on the problem of optimal power allocation for a two user interference channel characterized by mixed Channel State Information (CSI), which includes instantaneous information for the direct channels and statistical information for the interference channels. For this model, we introduce a general framework for optimizing the power allocation such as to maximize some generic Quality of Service (QoS) performance metric (or equivalently minimize some cost function). We model this problem as a function approximation problem where the function to be learned is the mapping between CSI and the solution to the optimization problem. We then tackle this problem borrowing ideas from ensemble learning. In particular, using generalized linear models (which are characterized by low complexity and can be implemented even at network nodes characterized by strong computational limitations), we produce different weak learners for learning to solve the considered optimization problem and based on ensemble learning theory, we combine such learners to produce stronger learners. We assess the performance of our framework by applying it on a particular resource allocation problem, and the obtained performance results indicate that the proposed approach can deliver near-optimal performance.","PeriodicalId":115090,"journal":{"name":"2021 International Balkan Conference on Communications and Networking (BalkanCom)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115137250","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：