Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976121
Huiying Zhou, Longqiang Wang, Baicun Wang, Geng Yang
Human-centric cyber physical system has become one of the most promising approaches in Industry 4.0. Humans are not obtained adequate attention in the traditional industrial process although humans serve as the orchestrator and beneficiary in the system. Human-centric physical system pertains to a safe and beneficial working environment, to the respect of human rights. This paper implements the inertial motion capture system into the industrial process, which supports to monitor human’s motion and recognize working activities regarding the assessment of the ergonomic performances. The inertial motion capture system integrates wearable inertial measurement units and the Unity3D application together for reconstructing human motion. Quaternion-based calibration algorithm is employed to achieve sensor-to-body segment alignment. Convolutional neural network based model classifies different activities of an assembly task when motion data are input of the deep learning network. The feasibility of the proposed system is validated by experiments.
{"title":"Human-centric Application in Cyber-Physical System: An Inertial-based Motion Capture and Recognition System","authors":"Huiying Zhou, Longqiang Wang, Baicun Wang, Geng Yang","doi":"10.1109/INDIN51773.2022.9976121","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976121","url":null,"abstract":"Human-centric cyber physical system has become one of the most promising approaches in Industry 4.0. Humans are not obtained adequate attention in the traditional industrial process although humans serve as the orchestrator and beneficiary in the system. Human-centric physical system pertains to a safe and beneficial working environment, to the respect of human rights. This paper implements the inertial motion capture system into the industrial process, which supports to monitor human’s motion and recognize working activities regarding the assessment of the ergonomic performances. The inertial motion capture system integrates wearable inertial measurement units and the Unity3D application together for reconstructing human motion. Quaternion-based calibration algorithm is employed to achieve sensor-to-body segment alignment. Convolutional neural network based model classifies different activities of an assembly task when motion data are input of the deep learning network. The feasibility of the proposed system is validated by experiments.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131149724","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976151
Vitor Vasconcelos, E. Leão, Natanael Ribeiro, F. Vasques, C. Montez
Wireless Sensor Networks (WSNs) based on the Long-Range radio modulation (LoRa) can use the LoRaWAN protocol as the medium access layer. However, this protocol only supports a single-hop star topology. As a consequence, devices can not use retransmissions along the network to extend their coverage area or to circumvent signal attenuation with distance, obstacles or interference from other radio sources. This paper proposes a multi-hop LoRaWAN wide-scale WSN based on a scheduled cluster-tree topology. This methodology can expand the spatial coverage of the network, decrease collisions, and improve overall network performance. A multi-hop cluster-tree topology eliminates the need for adjustments of LoRa radio parameters as an attempt to expand the single-hop coverage limitation. Simulation results show that the scheduled cluster-tree topology can scale the network coverage and significantly improve communication and energy consumption performances.
{"title":"A Scheduled Cluster-Tree Topology to Enable Wide-Scale LoRaWAN Networks","authors":"Vitor Vasconcelos, E. Leão, Natanael Ribeiro, F. Vasques, C. Montez","doi":"10.1109/INDIN51773.2022.9976151","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976151","url":null,"abstract":"Wireless Sensor Networks (WSNs) based on the Long-Range radio modulation (LoRa) can use the LoRaWAN protocol as the medium access layer. However, this protocol only supports a single-hop star topology. As a consequence, devices can not use retransmissions along the network to extend their coverage area or to circumvent signal attenuation with distance, obstacles or interference from other radio sources. This paper proposes a multi-hop LoRaWAN wide-scale WSN based on a scheduled cluster-tree topology. This methodology can expand the spatial coverage of the network, decrease collisions, and improve overall network performance. A multi-hop cluster-tree topology eliminates the need for adjustments of LoRa radio parameters as an attempt to expand the single-hop coverage limitation. Simulation results show that the scheduled cluster-tree topology can scale the network coverage and significantly improve communication and energy consumption performances.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131161335","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976180
Heping Pan
This paper walks through mathematical evolution of modern portfolio theory and multi-factor models and advances with a General Intelligent Portfolio Theory and underlying applications in stock markets. Following up the earlier form of the Intelligent Portfolio Theory, the new generalization extends in 3 dimensions: 1) three forms of intelligent portfolios – multi-asset multi-strategy, multi-strategy multi-asset and multi-trader; 2) strength investing with momentum rotation as an engine driving dynamic re-selection of assets or strategies or traders; 3) sector rotation in stock markets as a main form of strength investing and as a paradigm shift from diversification in portfolio theory. Applications in Chinese stock markets and international index futures are demonstrated with nontrivial performance achieved through testing on historical data.
{"title":"A General Intelligent Portfolio Theory with Strength Investing and Sector Rotation in Stock Markets","authors":"Heping Pan","doi":"10.1109/INDIN51773.2022.9976180","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976180","url":null,"abstract":"This paper walks through mathematical evolution of modern portfolio theory and multi-factor models and advances with a General Intelligent Portfolio Theory and underlying applications in stock markets. Following up the earlier form of the Intelligent Portfolio Theory, the new generalization extends in 3 dimensions: 1) three forms of intelligent portfolios – multi-asset multi-strategy, multi-strategy multi-asset and multi-trader; 2) strength investing with momentum rotation as an engine driving dynamic re-selection of assets or strategies or traders; 3) sector rotation in stock markets as a main form of strength investing and as a paradigm shift from diversification in portfolio theory. Applications in Chinese stock markets and international index futures are demonstrated with nontrivial performance achieved through testing on historical data.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125154012","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976068
Abdul Rehan Khan Mohammed, Jiayi Zhang, Ahmad Bilal
With the rapid growth in technology, the industries are fast-moving from the current automation standing into robotisation to increase productivity and deliver uniform quality. This requirement, in turn, has escalated the demand for robot control schemes. This paper proposes an observer-based robust adaptive tracking control scheme to minimise model uncertainties and external force disturbance effect to control the robot manipulator. No considerations are required for the upper bound of system uncertainties and disturbances in the control design. Plus, the speed of variation and the magnitude of unknown parameters and perturbations are assumed to have no limitations. The proposed control scheme uses an adaptation mechanism for a high gain nonlinear observer along with simplicity and universality properties to ensure robust tracking and make the system follow the desired reference model. Simulation results show that the proposed robust adaptive control scheme achieves boundedness for all the closed-loop signals and convergence of the tracking error.
{"title":"Observer-Based Robust Adaptive Tracking for Uncertain Robot Manipulators with External Force Disturbance Rejection","authors":"Abdul Rehan Khan Mohammed, Jiayi Zhang, Ahmad Bilal","doi":"10.1109/INDIN51773.2022.9976068","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976068","url":null,"abstract":"With the rapid growth in technology, the industries are fast-moving from the current automation standing into robotisation to increase productivity and deliver uniform quality. This requirement, in turn, has escalated the demand for robot control schemes. This paper proposes an observer-based robust adaptive tracking control scheme to minimise model uncertainties and external force disturbance effect to control the robot manipulator. No considerations are required for the upper bound of system uncertainties and disturbances in the control design. Plus, the speed of variation and the magnitude of unknown parameters and perturbations are assumed to have no limitations. The proposed control scheme uses an adaptation mechanism for a high gain nonlinear observer along with simplicity and universality properties to ensure robust tracking and make the system follow the desired reference model. Simulation results show that the proposed robust adaptive control scheme achieves boundedness for all the closed-loop signals and convergence of the tracking error.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121820392","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976132
Xiaoyue Liu, Cong Peng
Recently, deep transfer learning (TL) has successfully addressed the problem of fault diagnosis under variable operating conditions. Existing methods default that the source and target domains have the same label space, and solve distribution discrepancy problem under different working conditions by aligning their feature distributions. However, in the practical industry, is unlikely to guarantee the health conditions of the target domain data are consistent with the source domain. Therefore, industrial applications usually face the challenge of more difficult partial domain diagnosis scenarios. In this paper, a deep partial domain adaptation network based on a balanced alignment constraint strategy is proposed to realize cross-domain diagnosis. The proposed method combines balanced augmentation and subdomain alignment, which can effectively facilitate the positive transfer of shared categories. Meanwhile, the conditional entropy minimization is introduced to encourage the predictions of target domain samples with high confidence. The experimental results on the rolling bearing dataset verify the effectiveness and feasibility of the proposed method in handling the actual partial domain fault diagnosis problem.
{"title":"Partial Domain Intelligent Diagnosis Method for Rotor-Bearing System Based on Deep Learning","authors":"Xiaoyue Liu, Cong Peng","doi":"10.1109/INDIN51773.2022.9976132","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976132","url":null,"abstract":"Recently, deep transfer learning (TL) has successfully addressed the problem of fault diagnosis under variable operating conditions. Existing methods default that the source and target domains have the same label space, and solve distribution discrepancy problem under different working conditions by aligning their feature distributions. However, in the practical industry, is unlikely to guarantee the health conditions of the target domain data are consistent with the source domain. Therefore, industrial applications usually face the challenge of more difficult partial domain diagnosis scenarios. In this paper, a deep partial domain adaptation network based on a balanced alignment constraint strategy is proposed to realize cross-domain diagnosis. The proposed method combines balanced augmentation and subdomain alignment, which can effectively facilitate the positive transfer of shared categories. Meanwhile, the conditional entropy minimization is introduced to encourage the predictions of target domain samples with high confidence. The experimental results on the rolling bearing dataset verify the effectiveness and feasibility of the proposed method in handling the actual partial domain fault diagnosis problem.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127401206","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976136
S. Sajjadi, N. Bazmohammadi, A. Amani, M. Jalili, J. Guerrero, Xinghuo Yu
In this paper, control of Battery Storage Systems (BSS) in power distribution grids with residential consumers as well as prosumers equipped with rooftop photovoltaic (PV) solar panels and Electric Vehicles (EV) is addressed. Different features of these Distributed Energy Resources (DERs), such as intermittent behaviour and the difference between the maximum generation time and the maximum demand, have caused several issues for electricity distributors in delivering high quality power. Smart control and scheduling of ESS and EVs is a promising approach to protect the grid against extra power injection from prosumers during day times while the benefit of household owners from DERs are still achieved. In this context, the performance of model-based controllers such as model predictive controllers (MPC) is compared with model-free data driven controllers (DDC) considering different complex scenarios that may happen in a distribution grid. The control objective is to minimize the difference between the net power exchanged with the main grid from the estimated average net load of prosumers. Our study on the real consumption data of about 40 residential consumers/prosumers in Victoria, Australia, demonstrates the strength of data-driven control approaches to deal with the complex environment of power distribution grids in the presence of DERs.
{"title":"Control of Battery Storage Systems in Residential Grids: Model-based vs. Data-Driven Approaches","authors":"S. Sajjadi, N. Bazmohammadi, A. Amani, M. Jalili, J. Guerrero, Xinghuo Yu","doi":"10.1109/INDIN51773.2022.9976136","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976136","url":null,"abstract":"In this paper, control of Battery Storage Systems (BSS) in power distribution grids with residential consumers as well as prosumers equipped with rooftop photovoltaic (PV) solar panels and Electric Vehicles (EV) is addressed. Different features of these Distributed Energy Resources (DERs), such as intermittent behaviour and the difference between the maximum generation time and the maximum demand, have caused several issues for electricity distributors in delivering high quality power. Smart control and scheduling of ESS and EVs is a promising approach to protect the grid against extra power injection from prosumers during day times while the benefit of household owners from DERs are still achieved. In this context, the performance of model-based controllers such as model predictive controllers (MPC) is compared with model-free data driven controllers (DDC) considering different complex scenarios that may happen in a distribution grid. The control objective is to minimize the difference between the net power exchanged with the main grid from the estimated average net load of prosumers. Our study on the real consumption data of about 40 residential consumers/prosumers in Victoria, Australia, demonstrates the strength of data-driven control approaches to deal with the complex environment of power distribution grids in the presence of DERs.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127875936","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976103
M. Clausen, J. Schütz
The most vital requirement for the electric power system as a critical infrastructure is its security of supply. In course of the transition of the electric energy system, however, the security provided by the N-1 principle increasingly reaches its limits. The IT/OT convergence changes the threat structure significantly. New risk factors, that can lead to major blackouts, are added to the existing ones. The problem, however, the cost of security optimizations are not always in proportion to their value. Not every component is equally critical to the energy system, so the question arises, "How secure does my system need to be?". To adress the security-by-design principle, this contribution introduces a Security Metric (SecMet) that can be applied to Smart Grid architectures and its components and deliver an indicator for the "Securitisation Need" based on an individual risk assessment.
{"title":"Identifying Security Requirements for Smart Grid Components: A Smart Grid Security Metric","authors":"M. Clausen, J. Schütz","doi":"10.1109/INDIN51773.2022.9976103","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976103","url":null,"abstract":"The most vital requirement for the electric power system as a critical infrastructure is its security of supply. In course of the transition of the electric energy system, however, the security provided by the N-1 principle increasingly reaches its limits. The IT/OT convergence changes the threat structure significantly. New risk factors, that can lead to major blackouts, are added to the existing ones. The problem, however, the cost of security optimizations are not always in proportion to their value. Not every component is equally critical to the energy system, so the question arises, \"How secure does my system need to be?\". To adress the security-by-design principle, this contribution introduces a Security Metric (SecMet) that can be applied to Smart Grid architectures and its components and deliver an indicator for the \"Securitisation Need\" based on an individual risk assessment.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121194608","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976113
A. Silva, A. Simões, Renata Blanc
Collaborative robots are being increasingly used by manufacturing companies due to their potential to help companies cope with market volatility. Before introducing this technology, companies face the decision phase where they determine the investment feasibility. Decision models for cobot adoption can assist decision-makers in this task, but they require previous identification of decision criteria. Since existing literature overlooked this issue, this study aims to provide a list of decision criteria that can be considered in the cobot adoption decision process. These criteria were identified by a literature review of the benefits, advantages, and disadvantages of cobot adoption. Results show that flexibility, competitiveness, ergonomics, quality, safety, space, mobility, ease of programming, technical features, human-robot collaboration, and productivity are important aspects to consider when deciding whether to invest in cobots. The findings of this study provide a better understanding of the decision process for cobot adoption by listing decision criteria along with some indicators, which is an important input for the design of a decision-making process.
{"title":"Criteria to consider in a decision model for collaborative robot (cobot) adoption: A literature review","authors":"A. Silva, A. Simões, Renata Blanc","doi":"10.1109/INDIN51773.2022.9976113","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976113","url":null,"abstract":"Collaborative robots are being increasingly used by manufacturing companies due to their potential to help companies cope with market volatility. Before introducing this technology, companies face the decision phase where they determine the investment feasibility. Decision models for cobot adoption can assist decision-makers in this task, but they require previous identification of decision criteria. Since existing literature overlooked this issue, this study aims to provide a list of decision criteria that can be considered in the cobot adoption decision process. These criteria were identified by a literature review of the benefits, advantages, and disadvantages of cobot adoption. Results show that flexibility, competitiveness, ergonomics, quality, safety, space, mobility, ease of programming, technical features, human-robot collaboration, and productivity are important aspects to consider when deciding whether to invest in cobots. The findings of this study provide a better understanding of the decision process for cobot adoption by listing decision criteria along with some indicators, which is an important input for the design of a decision-making process.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115377198","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976100
Haiping Wang, Xing Zhou
Utilizing a large set of variables that include transaction information, public attention, blockchain information, macroeconomic variables and technical indicators, we compare different deep learning models with baseline methods, such as statistical and machine learning models, on Bitcoin volatility forecast. We find that feature selection approach strongly affects model performance. The results show that a simple Long Short-Term Memory (LSTM) model outperforms other models when using individual feature selection method.
{"title":"Less is More: Bitcoin Volatility Forecast Using Feature Selection and Deep Learning Models","authors":"Haiping Wang, Xing Zhou","doi":"10.1109/INDIN51773.2022.9976100","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976100","url":null,"abstract":"Utilizing a large set of variables that include transaction information, public attention, blockchain information, macroeconomic variables and technical indicators, we compare different deep learning models with baseline methods, such as statistical and machine learning models, on Bitcoin volatility forecast. We find that feature selection approach strongly affects model performance. The results show that a simple Long Short-Term Memory (LSTM) model outperforms other models when using individual feature selection method.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116023557","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976163
Mohammad Samadi Gharajeh, Tiago Carvalho, L. M. Pinho
Parallel programming models (e.g., OpenMP) are more and more used to improve the performance of real-time applications in modern processors. Nevertheless, these processors have complex architectures, being very difficult to understand their timing behavior. The main challenge with most of existing works is that they apply static timing analysis for simpler models or measurement-based analysis using traditional platforms (e.g., single core) or considering only sequential algorithms. How to provide an efficient configuration for the allocation of the parallel program in the computing units of the processor is still an open challenge. This paper studies the problem of performing timing analysis on complex multi-core platforms, pointing out a methodology to understand the applications’ timing behavior, and guide the configuration of the platform. As an example, the paper uses an OpenMP-based program of the Heat benchmark on a NVIDIA Jetson AGX Xavier. The main objectives are to analyze the execution time of OpenMP tasks, specify the best configuration of OpenMP directives, identify critical tasks, and discuss the predictability of the system/application. A Linux perf based measurement tool, which has been extended by our team, is applied to measure each task across multiple executions in terms of total CPU cycles, the number of cache accesses, and the number of cache misses at different cache levels, including L1, L2 and L3. The evaluation process is performed using the measurement of the performance metrics by our tool to study the predictability of the system/application.
{"title":"Configuration of Parallel Real-Time Applications on Multi-Core Processors","authors":"Mohammad Samadi Gharajeh, Tiago Carvalho, L. M. Pinho","doi":"10.1109/INDIN51773.2022.9976163","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976163","url":null,"abstract":"Parallel programming models (e.g., OpenMP) are more and more used to improve the performance of real-time applications in modern processors. Nevertheless, these processors have complex architectures, being very difficult to understand their timing behavior. The main challenge with most of existing works is that they apply static timing analysis for simpler models or measurement-based analysis using traditional platforms (e.g., single core) or considering only sequential algorithms. How to provide an efficient configuration for the allocation of the parallel program in the computing units of the processor is still an open challenge. This paper studies the problem of performing timing analysis on complex multi-core platforms, pointing out a methodology to understand the applications’ timing behavior, and guide the configuration of the platform. As an example, the paper uses an OpenMP-based program of the Heat benchmark on a NVIDIA Jetson AGX Xavier. The main objectives are to analyze the execution time of OpenMP tasks, specify the best configuration of OpenMP directives, identify critical tasks, and discuss the predictability of the system/application. A Linux perf based measurement tool, which has been extended by our team, is applied to measure each task across multiple executions in terms of total CPU cycles, the number of cache accesses, and the number of cache misses at different cache levels, including L1, L2 and L3. The evaluation process is performed using the measurement of the performance metrics by our tool to study the predictability of the system/application.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116225126","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：