Pub Date : 2020-10-30DOI: 10.1109/ICNSC48988.2020.9238050
Heng Liu, Xiaofen Zhang, J. Bi, Haitao Yuan, Mengchu Zhou
The ever-increasing deployment of cloud data centers causes high energy consumption, high cost, and harmful environmental pollution. To solve above problems, cloud service providers are actively exploring to use green cloud data centers (GCDCs) by using green energy. Yet it is challenging to accurately predict the future wind and solar energy before making intelligent task scheduling decisions. In addition, it is difficult to jointly optimize cost and revenue. In this work, to make optimal task scheduling, various types of applications, service level agreements, service rates, task loss probability, electricity prices and green energy in different GCDCs are considered. First, this work employs a long short-term memory network to predict wind and solar energy. Then, it adopts a bi-objective optimization algorithm to achieve a better trade-off between cost and revenue of GCDCs. Finally, it adopts real-world data including workload trace, wind energy, solar energy and electricity prices to demonstrate the effectiveness of the proposed energy prediction and task scheduling methods. It's shown that the proposed methods achieve higher performance than other neural network methods.
{"title":"Bi-objective Intelligent Task Scheduling for Green Clouds with Deep Learning-based Prediction","authors":"Heng Liu, Xiaofen Zhang, J. Bi, Haitao Yuan, Mengchu Zhou","doi":"10.1109/ICNSC48988.2020.9238050","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238050","url":null,"abstract":"The ever-increasing deployment of cloud data centers causes high energy consumption, high cost, and harmful environmental pollution. To solve above problems, cloud service providers are actively exploring to use green cloud data centers (GCDCs) by using green energy. Yet it is challenging to accurately predict the future wind and solar energy before making intelligent task scheduling decisions. In addition, it is difficult to jointly optimize cost and revenue. In this work, to make optimal task scheduling, various types of applications, service level agreements, service rates, task loss probability, electricity prices and green energy in different GCDCs are considered. First, this work employs a long short-term memory network to predict wind and solar energy. Then, it adopts a bi-objective optimization algorithm to achieve a better trade-off between cost and revenue of GCDCs. Finally, it adopts real-world data including workload trace, wind energy, solar energy and electricity prices to demonstrate the effectiveness of the proposed energy prediction and task scheduling methods. It's shown that the proposed methods achieve higher performance than other neural network methods.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127514833","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}
The track cleaning vehicle plays an important role in the cleaning and maintenance of the modern tramway. To better monitor the working conditions of it and eliminate the occurrence of accidents, a remote monitoring system is designed. Various types of sensors are installed on the track cleaning vehicle to collect the information and the data are transmitted to Tencent Cloud Database. The remote monitoring software based on QT helps administrators to monitor the working conditions of the track cleaning vehicle, providing warnings and safety tips to guarantee the normal operation of it by acquiring the data from Tencent Cloud Database. The experimental results show that administrators can remotely monitor the working conditions of the track cleaning vehicle by the system, which promotes the development of the track cleaning vehicle and provides a safety guarantee for the modern tram.
{"title":"Remote Monitoring System Of Track Cleaning Vehicles Based On 4G-Network","authors":"Zenan Lin, Yanming Huang, Yong-Jun Xie, Qiantong Wu, Xiaojie Huang, Jingui Li, Xin Liu","doi":"10.1109/ICNSC48988.2020.9238083","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238083","url":null,"abstract":"The track cleaning vehicle plays an important role in the cleaning and maintenance of the modern tramway. To better monitor the working conditions of it and eliminate the occurrence of accidents, a remote monitoring system is designed. Various types of sensors are installed on the track cleaning vehicle to collect the information and the data are transmitted to Tencent Cloud Database. The remote monitoring software based on QT helps administrators to monitor the working conditions of the track cleaning vehicle, providing warnings and safety tips to guarantee the normal operation of it by acquiring the data from Tencent Cloud Database. The experimental results show that administrators can remotely monitor the working conditions of the track cleaning vehicle by the system, which promotes the development of the track cleaning vehicle and provides a safety guarantee for the modern tram.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129192522","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}
Nowadays, AI education from Kindergarten to 9th grade(K-9) is in full swing in China, but many challenges exist, such as fragmented AI curricula, ineffective teaching tools, and uneven educational resources in urban and rural areas, etc. So this paper presents an in-depth study of the current status of K-9 AI education in Qingdao area through questionnaires, expert discussions, and field visits to sort out existing problems and then put forward corresponding advice and suggestions. Collected data is then analyzed from the perspectives of government, schools, teachers, students and parents, respectively. Results show that AI-related educational activities have been carried out in Qingdao area to improve students' AI literacy. But efforts are still needed to make such education more systematic, standardized, and personalized. Finally the paper made a list of recommendations corresponding to these needed efforts.
{"title":"K-9 Artificial Intelligence Education in Qingdao: Issues, Challenges and Suggestions","authors":"Xiaoyan Gong, Ying Tang, Xiwei Liu, Sifeng Jing, Wei Cui, Joleen Liang, Feiyue Wang","doi":"10.1109/ICNSC48988.2020.9238087","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238087","url":null,"abstract":"Nowadays, AI education from Kindergarten to 9th grade(K-9) is in full swing in China, but many challenges exist, such as fragmented AI curricula, ineffective teaching tools, and uneven educational resources in urban and rural areas, etc. So this paper presents an in-depth study of the current status of K-9 AI education in Qingdao area through questionnaires, expert discussions, and field visits to sort out existing problems and then put forward corresponding advice and suggestions. Collected data is then analyzed from the perspectives of government, schools, teachers, students and parents, respectively. Results show that AI-related educational activities have been carried out in Qingdao area to improve students' AI literacy. But efforts are still needed to make such education more systematic, standardized, and personalized. Finally the paper made a list of recommendations corresponding to these needed efforts.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116454639","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 : 2020-10-30DOI: 10.1109/ICNSC48988.2020.9238129
Shiyu Chen, Yanjie Li
Reinforcement learning (RL) is one of the popular methods for intelligent control and decision making in the field of robotics recently. The goal of RL is to learn an optimal policy of the agent by interacting with the environment via trail and error. There are two main algorithms for RL problems, including model-free and model-based methods. Model-free RL is driven by historical trajectories and empirical data of the agent to optimize the policy, which needs to take actions in the environment to collect the trajectory data and may cause the damage of the robot during training in the real environment. The main different between model-based and model-free RL is that a model of the transition probability in the interaction environment is employed. Thus the agent can search the optimal policy through internal simulation. However, the model of the transition probability is usually estimated from historical data in a single environment with statistical errors. Therefore, an issue is faced by the agent is that the optimal policy is sensitive to perturbations in the model of the environment which can lead to serious degradation in performance. Robust RL aims to learn a robust optimal policy that accounts for model uncertainty of the transition probability to systematically mitigate the sensitivity of the optimal policy in perturbed environments. In this overview, we begin with an introduction to the algorithms in RL, then focus on the model uncertainty of the transition probability in robust RL. In parallel, we highlight the current research and challenges of robust RL for robot control. To conclude, we describe some research areas in robust RL and look ahead to the future work about robot control in complex environments.
{"title":"An Overview of Robust Reinforcement Learning","authors":"Shiyu Chen, Yanjie Li","doi":"10.1109/ICNSC48988.2020.9238129","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238129","url":null,"abstract":"Reinforcement learning (RL) is one of the popular methods for intelligent control and decision making in the field of robotics recently. The goal of RL is to learn an optimal policy of the agent by interacting with the environment via trail and error. There are two main algorithms for RL problems, including model-free and model-based methods. Model-free RL is driven by historical trajectories and empirical data of the agent to optimize the policy, which needs to take actions in the environment to collect the trajectory data and may cause the damage of the robot during training in the real environment. The main different between model-based and model-free RL is that a model of the transition probability in the interaction environment is employed. Thus the agent can search the optimal policy through internal simulation. However, the model of the transition probability is usually estimated from historical data in a single environment with statistical errors. Therefore, an issue is faced by the agent is that the optimal policy is sensitive to perturbations in the model of the environment which can lead to serious degradation in performance. Robust RL aims to learn a robust optimal policy that accounts for model uncertainty of the transition probability to systematically mitigate the sensitivity of the optimal policy in perturbed environments. In this overview, we begin with an introduction to the algorithms in RL, then focus on the model uncertainty of the transition probability in robust RL. In parallel, we highlight the current research and challenges of robust RL for robot control. To conclude, we describe some research areas in robust RL and look ahead to the future work about robot control in complex environments.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131162994","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 : 2020-10-30DOI: 10.1109/ICNSC48988.2020.9238078
Liang Liu, Xiaopeng Luo, Zhangqing Zhu
In this paper, we propose the distributed formation control law to drive a group of unicycle-like vehicles to converge to a formation with a same orientation. The vehicles do not rely on a global coordinate frame. The network of the vehicles forms an acyclic digraph with no directed loops. We design the control law for vehicles without using position or direct distance measurements. Then we analyze the convergence and the properties of the closed-loop system. Finally, our simulation results certify the effectiveness of the proposed control laws.
{"title":"Distributed Formation Control of Unicycle-Like Vehicles Without Direct Distance Measurements","authors":"Liang Liu, Xiaopeng Luo, Zhangqing Zhu","doi":"10.1109/ICNSC48988.2020.9238078","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238078","url":null,"abstract":"In this paper, we propose the distributed formation control law to drive a group of unicycle-like vehicles to converge to a formation with a same orientation. The vehicles do not rely on a global coordinate frame. The network of the vehicles forms an acyclic digraph with no directed loops. We design the control law for vehicles without using position or direct distance measurements. Then we analyze the convergence and the properties of the closed-loop system. Finally, our simulation results certify the effectiveness of the proposed control laws.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"168 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131226525","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}
High-dimensional and sparse (HiDS) matrices generated by recommender systems (RSs) contain rich knowledge. A latent factor (LF) model can address such data effectively. Stochastic gradient descent (SGD) is an efficient algorithm for building a LF model on an HiDS matrix. However, it suffers slow convergence. To address this issue, this study proposes to implement a LF model with a proportional integral derivative (PID) controller. The main idea is to continuously apply a correction for SGD to accelerate the training process. Based on such design, a PID-based LF (PLF) model is proposed. Empirical studies on two HiDS matrices from RSs indicate that a PLF model outperforms an LF model in terms of both convergence rate and prediction accuracy for missing data.
{"title":"Accelerated Latent Factor Analysis for Recommender Systems via PID Controller","authors":"Jinli Li, Xuke Wu, Ye Yuan, Yajuan Wu, Kangkang Ma, Yue Zhou","doi":"10.1109/ICNSC48988.2020.9238055","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238055","url":null,"abstract":"High-dimensional and sparse (HiDS) matrices generated by recommender systems (RSs) contain rich knowledge. A latent factor (LF) model can address such data effectively. Stochastic gradient descent (SGD) is an efficient algorithm for building a LF model on an HiDS matrix. However, it suffers slow convergence. To address this issue, this study proposes to implement a LF model with a proportional integral derivative (PID) controller. The main idea is to continuously apply a correction for SGD to accelerate the training process. Based on such design, a PID-based LF (PLF) model is proposed. Empirical studies on two HiDS matrices from RSs indicate that a PLF model outperforms an LF model in terms of both convergence rate and prediction accuracy for missing data.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114567886","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 : 2020-10-30DOI: 10.1109/ICNSC48988.2020.9238112
Yinsen Jia, Yichen Li, Bo Xin, Chunlin Chen
In this paper, a strategy of path planning for autonomous obstacle avoidance using reinforcement learning for six-axis arms is proposed. This strategy gives priority to planning the obstacle avoidance path for the terminal of the mechanical arm, and then uses the calculated terminal path to plan the poses of the mechanical arm. For the points on the terminal path that the mechanical arm cannot avoid obstacles within the limit of the safe distance, this strategy will record these points as new obstacles and plan a new obstacle avoidance path for the terminal of mechanical arm. The above process is accelerated by the assisted learning strategies and looped until the correct path being calculated. The method proposed in this paper has been applied to a six-axis mechanical arm, and the simulation results show that this method can effectively plan an optimal path and poses for the mechanical arm.
{"title":"Path Planning with Autonomous Obstacle Avoidance Using Reinforcement Learning for Six-axis Arms","authors":"Yinsen Jia, Yichen Li, Bo Xin, Chunlin Chen","doi":"10.1109/ICNSC48988.2020.9238112","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238112","url":null,"abstract":"In this paper, a strategy of path planning for autonomous obstacle avoidance using reinforcement learning for six-axis arms is proposed. This strategy gives priority to planning the obstacle avoidance path for the terminal of the mechanical arm, and then uses the calculated terminal path to plan the poses of the mechanical arm. For the points on the terminal path that the mechanical arm cannot avoid obstacles within the limit of the safe distance, this strategy will record these points as new obstacles and plan a new obstacle avoidance path for the terminal of mechanical arm. The above process is accelerated by the assisted learning strategies and looped until the correct path being calculated. The method proposed in this paper has been applied to a six-axis mechanical arm, and the simulation results show that this method can effectively plan an optimal path and poses for the mechanical arm.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132962363","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 : 2020-10-30DOI: 10.1109/ICNSC48988.2020.9238111
Kosuke Nishizawa, M. Deng, Y. Noge
In this paper, a model of a microreactor unit using Peltier devices for cooling and design a control system is proposed. In detail, after describing the mathematical model of the microreactor in consideration of nonlinearity, a control system based on operator theory is designed using the proposed model. Next, the simulation results in the open-loop are shown, and the past model and the proposed model are compared. Finally, we show the simulation results of the proposed control system and confirm its effectiveness.
{"title":"Operator-based Nonlinear Modeling and Control for Microreactor","authors":"Kosuke Nishizawa, M. Deng, Y. Noge","doi":"10.1109/ICNSC48988.2020.9238111","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238111","url":null,"abstract":"In this paper, a model of a microreactor unit using Peltier devices for cooling and design a control system is proposed. In detail, after describing the mathematical model of the microreactor in consideration of nonlinearity, a control system based on operator theory is designed using the proposed model. Next, the simulation results in the open-loop are shown, and the past model and the proposed model are compared. Finally, we show the simulation results of the proposed control system and confirm its effectiveness.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129249012","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 : 2020-10-30DOI: 10.1109/ICNSC48988.2020.9238113
Guang Jin, M. Deng
In this paper, a new nonlinear forced vibration control scheme using an operator-based robust right coprime factorization approach is considered for forced vibration control on a flexible plate with piezoelectric actuator. First, for considering the effect of hysteresis nonlinearity from the piezoelectric actuator, the Prandtl-Ishlinskii (P-I) hysteresis model is used to describe it. Also, a dynamic model of flexible plate is given by the theory of thin plates. For guaranteeing the robust stability of the nonin-ear forced vibration control system, operator-based controllers are designed. Simultaneously, for improving forced vibration control performance, the time-varying unimodular function is constructed by the designed controllers. If the inverse of the time-varying unimodular function tends to zero by the operator-based controllers and designed compensator, the output can be made arbitrarily small. Finally, the effectiveness of the proposed nonlinear control system is confirmed by simulation results.
{"title":"Time-varying unimodular function based robust right coprime factorization for nonlinear forced vibration control system","authors":"Guang Jin, M. Deng","doi":"10.1109/ICNSC48988.2020.9238113","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238113","url":null,"abstract":"In this paper, a new nonlinear forced vibration control scheme using an operator-based robust right coprime factorization approach is considered for forced vibration control on a flexible plate with piezoelectric actuator. First, for considering the effect of hysteresis nonlinearity from the piezoelectric actuator, the Prandtl-Ishlinskii (P-I) hysteresis model is used to describe it. Also, a dynamic model of flexible plate is given by the theory of thin plates. For guaranteeing the robust stability of the nonin-ear forced vibration control system, operator-based controllers are designed. Simultaneously, for improving forced vibration control performance, the time-varying unimodular function is constructed by the designed controllers. If the inverse of the time-varying unimodular function tends to zero by the operator-based controllers and designed compensator, the output can be made arbitrarily small. Finally, the effectiveness of the proposed nonlinear control system is confirmed by simulation results.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127777927","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 : 2020-10-30DOI: 10.1109/ICNSC48988.2020.9238084
Lingyu Zhang, Jiabao Zhao, Min Zhang
With the development of Internet technology, IT systems are getting more and more complex, in which there are two main relationships among system components: service call relationship and deployment configuration relationship. Once a local anomaly occurs in the system, it tends to spread, triggering emergent and dense concurrent alarms. Hence, it is important to quickly and precisely locate the root cause of concurrent alarms. In this paper, we first construct an anomaly propagation graph using collected system data. Then, based on the graph, we propose two optional algorithms: random walk and state iteration, to track anomaly propagation process and locate the root cause. Simulation experiments demonstrate that our proposed method can localize root causes correctly and rapidly for scenarios with complex call chains and resource competition, and is robust to alarm error. The proposed method pays more attention to system characteristics and depends little on experience knowledge of IT operators.
{"title":"Root Cause Analysis of Concurrent Alarms Based on Random Walk over Anomaly Propagation Graph","authors":"Lingyu Zhang, Jiabao Zhao, Min Zhang","doi":"10.1109/ICNSC48988.2020.9238084","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238084","url":null,"abstract":"With the development of Internet technology, IT systems are getting more and more complex, in which there are two main relationships among system components: service call relationship and deployment configuration relationship. Once a local anomaly occurs in the system, it tends to spread, triggering emergent and dense concurrent alarms. Hence, it is important to quickly and precisely locate the root cause of concurrent alarms. In this paper, we first construct an anomaly propagation graph using collected system data. Then, based on the graph, we propose two optional algorithms: random walk and state iteration, to track anomaly propagation process and locate the root cause. Simulation experiments demonstrate that our proposed method can localize root causes correctly and rapidly for scenarios with complex call chains and resource competition, and is robust to alarm error. The proposed method pays more attention to system characteristics and depends little on experience knowledge of IT operators.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"257 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115795176","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：