Pub Date : 2020-10-30DOI: 10.1109/ICNSC48988.2020.9238127
Xucheng Huang, Jie Jiang, Shuanghua Yang, Yulong Ding
LoRa is a low-power long-range network technology, which is used widely in power sensitive and maintenance free Internet of Things applications. LoRa only defines the physical layer protocol, while LoRaWAN is a medium access control (MAC) layer protocol above it. However, simply using ALOHA in LoRaWAN makes a high package collision rate when the number of the end-devices in the network is large, since many end-devices will send the packages to gateway at the same time. To solve this, we present a reinforcement learning (RL) based multi access method for LoRaWAN, which allows end-devices decide when to transmit data based on the environment and reduce the package collision rate. A comparation between the RL method and ALOHA is also included in the paper, which shows that the RL method has a lower package collision rate.
{"title":"A Reinforcement Learning Based Medium Access Control Method for LoRa Networks","authors":"Xucheng Huang, Jie Jiang, Shuanghua Yang, Yulong Ding","doi":"10.1109/ICNSC48988.2020.9238127","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238127","url":null,"abstract":"LoRa is a low-power long-range network technology, which is used widely in power sensitive and maintenance free Internet of Things applications. LoRa only defines the physical layer protocol, while LoRaWAN is a medium access control (MAC) layer protocol above it. However, simply using ALOHA in LoRaWAN makes a high package collision rate when the number of the end-devices in the network is large, since many end-devices will send the packages to gateway at the same time. To solve this, we present a reinforcement learning (RL) based multi access method for LoRaWAN, which allows end-devices decide when to transmit data based on the environment and reduce the package collision rate. A comparation between the RL method and ALOHA is also included in the paper, which shows that the RL method has a lower package collision rate.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"4 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":"114093234","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/ICNSC48988.2020.9238120
Lifeng Liu, Bo Xin, Zhangqing Zhu, Tao Jiang, Chunlin Chen, Tao Wu
We evaluate the density of rape pictures based on Convolutional Networks, and compare methods via fused features combined with two kinds of regression approaches: Support Vector Regression, SVR, and Lasso Regression. The Convolutional Networks extract the features of rape images through convolutional layers, pooling layers and activation functions, and then, fully connected layers regress the extracted features to the density value. The fused features involve three types of features: image energy, local binary pattern(LBP) features and Gabor wavelets texture features. First, the method extracts the fused features through python scikit-learn packages [1], and then regression model regresses the fused features to the density value by Support Vector Regression [2] [3] or Lasso Regression [4].
{"title":"Density Evaluation based on Convolutional Networks in Rape Images","authors":"Lifeng Liu, Bo Xin, Zhangqing Zhu, Tao Jiang, Chunlin Chen, Tao Wu","doi":"10.1109/ICNSC48988.2020.9238120","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238120","url":null,"abstract":"We evaluate the density of rape pictures based on Convolutional Networks, and compare methods via fused features combined with two kinds of regression approaches: Support Vector Regression, SVR, and Lasso Regression. The Convolutional Networks extract the features of rape images through convolutional layers, pooling layers and activation functions, and then, fully connected layers regress the extracted features to the density value. The fused features involve three types of features: image energy, local binary pattern(LBP) features and Gabor wavelets texture features. First, the method extracts the fused features through python scikit-learn packages [1], and then regression model regresses the fused features to the density value by Support Vector Regression [2] [3] or Lasso Regression [4].","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"7 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":"121307343","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/ICNSC48988.2020.9238130
M. Jin, Jianhong Ye, Jiliang Luo, Yan Lin
Predictive monitoring is a branch of process mining to provide some valuable information that enables proactive corrective actions to mitigate risks. This paper proposes a new predictive monitoring algorithm, which divides the data processing into three parts: prefix extraction, prefix bucketing and prefix encoding. The presented encoding methods are based on the data structure of event log, and it will cause the loss of information in the raw data. Our main contribution is to define a new global feature encoding method, which keeps more information in raw data and has better scalability. Experiments are presented to demonstrate the proposed approach.
{"title":"Predictive Monitoring Algorithm Based on Global Feature Encoding","authors":"M. Jin, Jianhong Ye, Jiliang Luo, Yan Lin","doi":"10.1109/ICNSC48988.2020.9238130","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238130","url":null,"abstract":"Predictive monitoring is a branch of process mining to provide some valuable information that enables proactive corrective actions to mitigate risks. This paper proposes a new predictive monitoring algorithm, which divides the data processing into three parts: prefix extraction, prefix bucketing and prefix encoding. The presented encoding methods are based on the data structure of event log, and it will cause the loss of information in the raw data. Our main contribution is to define a new global feature encoding method, which keeps more information in raw data and has better scalability. Experiments are presented to demonstrate the proposed approach.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"20 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":"123770132","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/ICNSC48988.2020.9238076
Zhan Deng, Jianyu Wang
Dempster Shafer evidence theory is widely used in the field of information fusion. However, when there is a high conflict between the evidence, Dempster Shafer fusion method will generate a counter intuitive result. To address this issue, by considering the credibility and uncertainty information of the evidence, we propose a new multi-sensor data fusion method based on Hellinger distance and belief entropy. The new multi-sensor data fusion method consists of three main procedures. Firstly, the probability transformation method is used to transform the basic probability assignment into the probability distribution, then the Hellinger distance is utilized to measure the distance between the evidence, and the credibility of the evidence is calculated by the distance between the evidence. Secondly, considering the information volume of the evidence. In this paper, belief entropy is applied to measure the information volume of the evidence, and then the information volume of the evidence is used to modify the credibility of the evidence. Finally, the credibility of the evidence is taken as a weight factor to modify the original evidence to obtain the weighted average evidence, and then the weighted average evidence is fused with Dempster Shafer combination rule to achieve the final fusion result. Numerical examples and fault diagnosis applications illustrate the effectiveness and accuracy of the proposed method.
{"title":"Conflicting evidence combination method based on evidence distance and belief entropy","authors":"Zhan Deng, Jianyu Wang","doi":"10.1109/ICNSC48988.2020.9238076","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238076","url":null,"abstract":"Dempster Shafer evidence theory is widely used in the field of information fusion. However, when there is a high conflict between the evidence, Dempster Shafer fusion method will generate a counter intuitive result. To address this issue, by considering the credibility and uncertainty information of the evidence, we propose a new multi-sensor data fusion method based on Hellinger distance and belief entropy. The new multi-sensor data fusion method consists of three main procedures. Firstly, the probability transformation method is used to transform the basic probability assignment into the probability distribution, then the Hellinger distance is utilized to measure the distance between the evidence, and the credibility of the evidence is calculated by the distance between the evidence. Secondly, considering the information volume of the evidence. In this paper, belief entropy is applied to measure the information volume of the evidence, and then the information volume of the evidence is used to modify the credibility of the evidence. Finally, the credibility of the evidence is taken as a weight factor to modify the original evidence to obtain the weighted average evidence, and then the weighted average evidence is fused with Dempster Shafer combination rule to achieve the final fusion result. Numerical examples and fault diagnosis applications illustrate the effectiveness and accuracy of the proposed method.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"10 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":"116953615","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/ICNSC48988.2020.9238072
Zihui Wang, Zhi Wang, Chunlin Chen
Sparse interaction in multiagent tasks is an important approach to reduce the exponential computational cost for multiagent reinforcement learning (MARL) systems. How to select proper equilibrium solutions is the key to find the optimal policy and to improve the learning performance when collisions occur. We propose a new MARL algorithm, Efficient Coordination based MARL with Sparse Interactions (ECoSI), using the sparse interaction framework and an efficient coordination mechanism, where equilibrium solutions are selected via Nash equilibrium and Chicken game. ECoSI not only separates the Q-value updating rule in joint states from non-joint states with sparse interactions to achieve lower computation and storage complexity, but also takes advantage of efficient coordination with equilibrium solutions to find the optimal policy. Experimental results demonstrate the effectiveness and robustness of ECoSI compared to other state-of-the-art MARL algorithms.
多智能体任务中的稀疏交互是降低多智能体强化学习(MARL)系统指数计算成本的重要途径。如何选择合适的平衡解是在发生碰撞时找到最优策略和提高学习性能的关键。本文提出了一种新的MARL算法——基于稀疏交互的高效协调(Efficient Coordination based MARL with Sparse Interactions, ECoSI)，该算法利用稀疏交互框架和高效协调机制，通过纳什均衡和Chicken game选择均衡解。ECoSI不仅将联合状态下的q值更新规则从相互作用稀疏的非联合状态中分离出来，降低了计算和存储复杂度，而且利用与平衡解的有效协调来寻找最优策略。实验结果证明了ECoSI与其他先进的MARL算法相比的有效性和鲁棒性。
{"title":"A Coordinated Multiagent Reinforcement Learning Method Using Chicken Game","authors":"Zihui Wang, Zhi Wang, Chunlin Chen","doi":"10.1109/ICNSC48988.2020.9238072","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238072","url":null,"abstract":"Sparse interaction in multiagent tasks is an important approach to reduce the exponential computational cost for multiagent reinforcement learning (MARL) systems. How to select proper equilibrium solutions is the key to find the optimal policy and to improve the learning performance when collisions occur. We propose a new MARL algorithm, Efficient Coordination based MARL with Sparse Interactions (ECoSI), using the sparse interaction framework and an efficient coordination mechanism, where equilibrium solutions are selected via Nash equilibrium and Chicken game. ECoSI not only separates the Q-value updating rule in joint states from non-joint states with sparse interactions to achieve lower computation and storage complexity, but also takes advantage of efficient coordination with equilibrium solutions to find the optimal policy. Experimental results demonstrate the effectiveness and robustness of ECoSI compared to other state-of-the-art MARL algorithms.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"29 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":"115291311","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/ICNSC48988.2020.9238094
Yunpeng Zhou, Zhangqing Zhu, Bo Xin
We propose a one-step Local Feature Extraction Network framework to solve the sparse feature matching problem. In our network, we use raw camera data and the Structure from Motion (SfM) algorithm to restore the corresponding relationships of the different feature map. Our network combines the detector and descriptor as one step to build an end-to-end Local Feature Extraction network. At the same time, the whole process is differentiable and we train our network by the loss of feature map. Finally, we train our network on indoor datasets and prove its accuracy and rapidity advantage over other methods.
{"title":"One-step Local Feature Extraction using CNN","authors":"Yunpeng Zhou, Zhangqing Zhu, Bo Xin","doi":"10.1109/ICNSC48988.2020.9238094","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238094","url":null,"abstract":"We propose a one-step Local Feature Extraction Network framework to solve the sparse feature matching problem. In our network, we use raw camera data and the Structure from Motion (SfM) algorithm to restore the corresponding relationships of the different feature map. Our network combines the detector and descriptor as one step to build an end-to-end Local Feature Extraction network. At the same time, the whole process is differentiable and we train our network by the loss of feature map. Finally, we train our network on indoor datasets and prove its accuracy and rapidity advantage over other methods.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"13 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":"133141099","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}
This paper proposes an end-to-end deep reinforcement learning approach for mobile robot navigation with dynamic obstacles avoidance. Using experience collected in a simulation environment, a convolutional neural network (CNN) is trained to predict proper steering actions of a robot from its egocentric local occupancy maps, which accommodate various sensors and fusion algorithms. The trained neural network is then transferred and executed on a real-world mobile robot to guide its local path planning. The new approach is evaluated both qualitatively and quantitatively in simulation and realworld robot experiments. The results show that the map-based end-to-end navigation model is easy to be deployed to a robotic platform, robust to sensor noise and outperforms other existing DRL-based models in many indicators.
{"title":"Robot Navigation with Map-Based Deep Reinforcement Learning","authors":"Guangda Chen, Lifan Pan, Yu'an Chen, Pei Xu, Zhiqiang Wang, Peichen Wu, Jianmin Ji, Xiaoping Chen","doi":"10.1109/ICNSC48988.2020.9238090","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238090","url":null,"abstract":"This paper proposes an end-to-end deep reinforcement learning approach for mobile robot navigation with dynamic obstacles avoidance. Using experience collected in a simulation environment, a convolutional neural network (CNN) is trained to predict proper steering actions of a robot from its egocentric local occupancy maps, which accommodate various sensors and fusion algorithms. The trained neural network is then transferred and executed on a real-world mobile robot to guide its local path planning. The new approach is evaluated both qualitatively and quantitatively in simulation and realworld robot experiments. The results show that the map-based end-to-end navigation model is easy to be deployed to a robotic platform, robust to sensor noise and outperforms other existing DRL-based models in many indicators.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130051561","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 : 2020-01-07DOI: 10.1109/ICNSC48988.2020.9238119
Xiaoran Yan, O. Sporns, Andrea Avena-Koenigsberger
We propose a information theoretical framework to capture transition and information costs of network navigation models. Based on the minimum description length principle and the Markov decision process, we demonstrate that efficient global navigation can be achieved with only partial information. Additionally, we derived a scalable algorithm for optimal solutions under certain conditions. The proposed algorithm can be interpreted as a dynamical process on network, making it a useful tool for analysing and understanding navigation strategies on real world networks.
{"title":"Efficient network navigation with partial information","authors":"Xiaoran Yan, O. Sporns, Andrea Avena-Koenigsberger","doi":"10.1109/ICNSC48988.2020.9238119","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238119","url":null,"abstract":"We propose a information theoretical framework to capture transition and information costs of network navigation models. Based on the minimum description length principle and the Markov decision process, we demonstrate that efficient global navigation can be achieved with only partial information. Additionally, we derived a scalable algorithm for optimal solutions under certain conditions. The proposed algorithm can be interpreted as a dynamical process on network, making it a useful tool for analysing and understanding navigation strategies on real world networks.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124490701","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 : 2019-11-29DOI: 10.1109/ICNSC48988.2020.9238054
Taichao Li, Min Jiang
In this paper, we propose one new quantum key agreement (QKA) protocol using non-maximally entangled Bell states with positive operator-valued measurement (POVM). It is designed for multi-party QKA by non-maximally entangled Bell states with POVM. Since Bell states and single particle can be obtained by various physical systems, thus, our protocol is feasible based on the current technology. It is secure against the outsider and participant attack. Further, it is shown that the shared key is decided by all participants. Therefore, it could guarantee the security and fairness.
{"title":"Quantum key agreement via non-maximally entangled Bell states","authors":"Taichao Li, Min Jiang","doi":"10.1109/ICNSC48988.2020.9238054","DOIUrl":"https://doi.org/10.1109/ICNSC48988.2020.9238054","url":null,"abstract":"In this paper, we propose one new quantum key agreement (QKA) protocol using non-maximally entangled Bell states with positive operator-valued measurement (POVM). It is designed for multi-party QKA by non-maximally entangled Bell states with POVM. Since Bell states and single particle can be obtained by various physical systems, thus, our protocol is feasible based on the current technology. It is secure against the outsider and participant attack. Further, it is shown that the shared key is decided by all participants. Therefore, it could guarantee the security and fairness.","PeriodicalId":412290,"journal":{"name":"2020 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131171752","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：