Pub Date : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674080
H. Monday, J. Li, G. Nneji, A. Z. Yutra, Bona D. Lemessa, Saifun Nahar, E. James, A. Haq
The electrical system's dependability, security, and efficiency are all improved through smart grid technologies. Its dependence on digital communication technology, on the other hand, introduces new risks and vulnerabilities that should be examined for the purpose to providing effective and trustworthy service delivery. This study presents a method for the detection of distributed denial of service (DDoS) attacks on smart grid infrastructure. Continuous wavelet transform (CWT) is used in the suggested approach to convert one-dimensional traffic data to two-dimensional time-frequency domain scalogram as the input to the wavelet convolutional neural network (WavCovNet) to detect anomalous behavior in the data by distinguishing attack features from normal patterns. Our results demonstrate that the proposed approach detects DDoS attacks with a high rate of detection and with a very low rate of false alarm.
{"title":"The Capability of Wavelet Convolutional Neural Network for Detecting Cyber Attack of Distributed Denial of Service in Smart Grid","authors":"H. Monday, J. Li, G. Nneji, A. Z. Yutra, Bona D. Lemessa, Saifun Nahar, E. James, A. Haq","doi":"10.1109/ICCWAMTIP53232.2021.9674080","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674080","url":null,"abstract":"The electrical system's dependability, security, and efficiency are all improved through smart grid technologies. Its dependence on digital communication technology, on the other hand, introduces new risks and vulnerabilities that should be examined for the purpose to providing effective and trustworthy service delivery. This study presents a method for the detection of distributed denial of service (DDoS) attacks on smart grid infrastructure. Continuous wavelet transform (CWT) is used in the suggested approach to convert one-dimensional traffic data to two-dimensional time-frequency domain scalogram as the input to the wavelet convolutional neural network (WavCovNet) to detect anomalous behavior in the data by distinguishing attack features from normal patterns. Our results demonstrate that the proposed approach detects DDoS attacks with a high rate of detection and with a very low rate of false alarm.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130248723","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 : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674132
Dai Xuanhui, Chen Juan, Wen Quan
Network pruning is widely used for compressing large neural networks to save computational resources. In traditional pruning methods, predefined hyperparameters are often required to determine the network structure of the target small network. However, too many hyperparameters are often undesirable. Therefore, we use the transformable architecture search (TAS) method to dynamically search the network structure of each layer when pruning the network width. In the TAS method, the channels number of the pruned network in each layer is represented by a learnable probability distribution. By minimizing computation cost, the probability distribution can be calculated and used to get the width configuration of the target pruned network. Then, the depth of the network was compressed based on the model get in the previous step. The method for compressing depth is block-wise intermediate representation training. This method is based on the hint training, where the network depth is compressed by comparing the intermediate representation of each layer of two equally wide teacher and student models. In the experiments, about 0.4% improvement over the existing method was viewed for the ResNet network on both CIFAR10 and CIFAR100 datasets.
{"title":"Network Pruning Based On Architecture Search and Intermediate Representation","authors":"Dai Xuanhui, Chen Juan, Wen Quan","doi":"10.1109/ICCWAMTIP53232.2021.9674132","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674132","url":null,"abstract":"Network pruning is widely used for compressing large neural networks to save computational resources. In traditional pruning methods, predefined hyperparameters are often required to determine the network structure of the target small network. However, too many hyperparameters are often undesirable. Therefore, we use the transformable architecture search (TAS) method to dynamically search the network structure of each layer when pruning the network width. In the TAS method, the channels number of the pruned network in each layer is represented by a learnable probability distribution. By minimizing computation cost, the probability distribution can be calculated and used to get the width configuration of the target pruned network. Then, the depth of the network was compressed based on the model get in the previous step. The method for compressing depth is block-wise intermediate representation training. This method is based on the hint training, where the network depth is compressed by comparing the intermediate representation of each layer of two equally wide teacher and student models. In the experiments, about 0.4% improvement over the existing method was viewed for the ResNet network on both CIFAR10 and CIFAR100 datasets.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131427718","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 : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674134
Yao Siyi, Xu Jianliang, Shen Weiguo, Wang Li, Li Wanchun
In this paper, the multi-target positioning research is carried out on the basis of a passive multi-sensor reconnaissance system, which requires an algorithm to solve the problem of multi-target data association and positioning in the presence of false alarms and missed detections (also named non-ideal). This paper proposes a new data association and localization algorithm. This new algorithm adopts the method of hierarchical association of data, achieves data association through coarse association and fine association, and finally deletes redundant targets through backtracking. The simulation results show the effectiveness and stability of the algorithm.
{"title":"Research On Multi - Target Data Association and Location Algorithm Based On Passive Multi - Sensor System","authors":"Yao Siyi, Xu Jianliang, Shen Weiguo, Wang Li, Li Wanchun","doi":"10.1109/ICCWAMTIP53232.2021.9674134","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674134","url":null,"abstract":"In this paper, the multi-target positioning research is carried out on the basis of a passive multi-sensor reconnaissance system, which requires an algorithm to solve the problem of multi-target data association and positioning in the presence of false alarms and missed detections (also named non-ideal). This paper proposes a new data association and localization algorithm. This new algorithm adopts the method of hierarchical association of data, achieves data association through coarse association and fine association, and finally deletes redundant targets through backtracking. The simulation results show the effectiveness and stability of the algorithm.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133872210","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 : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674140
Zhang Jianfang, Zhang Yaqi, Suyan Chen
The relationship between blockchain and emergency supply chain coordination is analyzed, and the utility model of emergency supply chain coordination is established. Applying the characteristics of blockchain technology, the emergency logistics activity process is recorded on the blockchain. When the emergency logistics activity changes, the emergency supply chain members can record the emergency logistics activity changes to ensure that the emergency logistics activity process is open and transparent. Combined with the blockchain smart contract algorithm of emergency supply chain logistics activities, a blockchain-based emergency supply chain synergy model is constructed.
{"title":"Research On Collaborative Management Model of Emergency Supply Chain Based On Blockchain","authors":"Zhang Jianfang, Zhang Yaqi, Suyan Chen","doi":"10.1109/ICCWAMTIP53232.2021.9674140","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674140","url":null,"abstract":"The relationship between blockchain and emergency supply chain coordination is analyzed, and the utility model of emergency supply chain coordination is established. Applying the characteristics of blockchain technology, the emergency logistics activity process is recorded on the blockchain. When the emergency logistics activity changes, the emergency supply chain members can record the emergency logistics activity changes to ensure that the emergency logistics activity process is open and transparent. Combined with the blockchain smart contract algorithm of emergency supply chain logistics activities, a blockchain-based emergency supply chain synergy model is constructed.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131692849","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 : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674075
Di Liu, Zhizhao Feng, Zhao Du
Traditional network fault detection methods need to collect data for training, which has data security problems. In recent years, as people pay more and more attention to data privacy, how to ensure data security has become more and more important. At the same time, because the network fault detection needs to meet certain real-time requirements, how to improve the detection speed is also an urgent problem to be solved. Based on the above two problems, this paper proposes a network fault detection algorithm DNFD-SRU based on federated learning and SRU. Federated learning can train the model on the premise of ensuring data security, and SRU has faster training speed.
{"title":"DNFD-SRU: A Distributed Network Fault Detection Method Based on SRU","authors":"Di Liu, Zhizhao Feng, Zhao Du","doi":"10.1109/ICCWAMTIP53232.2021.9674075","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674075","url":null,"abstract":"Traditional network fault detection methods need to collect data for training, which has data security problems. In recent years, as people pay more and more attention to data privacy, how to ensure data security has become more and more important. At the same time, because the network fault detection needs to meet certain real-time requirements, how to improve the detection speed is also an urgent problem to be solved. Based on the above two problems, this paper proposes a network fault detection algorithm DNFD-SRU based on federated learning and SRU. Federated learning can train the model on the premise of ensuring data security, and SRU has faster training speed.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129870320","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 : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674124
Muhammad Hanif Tunio, Liao Jianping, Muhammad Hassaan Farooq Butt, Imran Memon
The Rice crop is considered one of the most widely grown crops in Asia and it is susceptible to various types of illnesses at different stages of production. Food safety and production can be affected by rice plant diseases, as well as a significant decline in the quality and quantity of agricultural goods. Plant diseases can potentially prevent grain harvesting entirely in severe circumstances. As a result, automation of identification and diagnosis of plant disease is widely needed in the agriculture field. Many approaches for doing this problem have been offered with deep learning rising as the preferred method because of its excellent achievement. In this proposed research, we used Hybrid deep CNN transfer learning with rice plant images or the classification and identification of various rice diseases, we employed Transfer Learning to generate our deep learning model using Rice_Leaf_Dataset from a secondary source. The proposed model is 90.8% accurate, Experiments show that the proposed approach is viable, and it can be used to detect plant diseases efficiently and outperformed.
{"title":"Identification and Classification of Rice Plant Disease Using Hybrid Transfer Learning","authors":"Muhammad Hanif Tunio, Liao Jianping, Muhammad Hassaan Farooq Butt, Imran Memon","doi":"10.1109/ICCWAMTIP53232.2021.9674124","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674124","url":null,"abstract":"The Rice crop is considered one of the most widely grown crops in Asia and it is susceptible to various types of illnesses at different stages of production. Food safety and production can be affected by rice plant diseases, as well as a significant decline in the quality and quantity of agricultural goods. Plant diseases can potentially prevent grain harvesting entirely in severe circumstances. As a result, automation of identification and diagnosis of plant disease is widely needed in the agriculture field. Many approaches for doing this problem have been offered with deep learning rising as the preferred method because of its excellent achievement. In this proposed research, we used Hybrid deep CNN transfer learning with rice plant images or the classification and identification of various rice diseases, we employed Transfer Learning to generate our deep learning model using Rice_Leaf_Dataset from a secondary source. The proposed model is 90.8% accurate, Experiments show that the proposed approach is viable, and it can be used to detect plant diseases efficiently and outperformed.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130765672","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 : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674170
Zhu Ya Lou, L. Fan, Kuang Ping, Feng Dong
The Actor Relation Graph (ARG) is an effective method for detecting group behaviour but still needs improvement in some areas. In this paper, we propose using the sum of absolute differences (SAD) to compute the similarity of characters' appearance, introduce deep mutual learning to support the network's training, and add a visualization model. By training with the extended dataset, the results show that our improved network can achieve the expected better prediction accuracy of group activities.
{"title":"Mutual Learning Networks Of Actor Relation Graph For Group Activity Recognition","authors":"Zhu Ya Lou, L. Fan, Kuang Ping, Feng Dong","doi":"10.1109/ICCWAMTIP53232.2021.9674170","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674170","url":null,"abstract":"The Actor Relation Graph (ARG) is an effective method for detecting group behaviour but still needs improvement in some areas. In this paper, we propose using the sum of absolute differences (SAD) to compute the similarity of characters' appearance, introduce deep mutual learning to support the network's training, and add a visualization model. By training with the extended dataset, the results show that our improved network can achieve the expected better prediction accuracy of group activities.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"169 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126647059","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 : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674054
Chen Hongle, Shi Qirui, Chen Juan, Wen Quan
A popular method for shrinking over-parameterized networks nowadays is pruning, which can efficiently reduce the number of computational parameters and computational cost of the network and has almost the same high accuracy as the original network. The general weighted pruning algorithm can only reduce the number of parameters based on the original network structure, but cannot reduce the width and depth of the pruned network. While the knowledge distillation algorithm can solve the problem by compressing the network structure, it cannot make further modifications on the processed network. To further reduce the network structure, we propose a model compression algorithm, HKDP, a hybrid method combining knowledge distillation and network pruning that can significantly reduce the overall size of the network and maintain substantial accuracy. This approach obtains the advantages of knowledge distillation and pruning, which achieves 10 times higher compression rate and 2 percent higher accuracy than using either algorithm alone. Concretely, we apply a stage-wise knowledge distillation algorithm in the front that can quickly and efficiently reduce the original model structure; we also apply a Stochastic Gradient Descent (SGD) based pruning method and introduce the concept of global sparsity, which allows us to customize the compression rate of the model. Our experiments on CIFAR-10 and MNIST show that our hybrid optimization algorithm has higher model accuracy and model compression ratio compared to other competitors' network compression algorithms.
{"title":"HKDP: A Hybrid Approach On Knowledge Distillation and Pruning for Neural Network Compression","authors":"Chen Hongle, Shi Qirui, Chen Juan, Wen Quan","doi":"10.1109/ICCWAMTIP53232.2021.9674054","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674054","url":null,"abstract":"A popular method for shrinking over-parameterized networks nowadays is pruning, which can efficiently reduce the number of computational parameters and computational cost of the network and has almost the same high accuracy as the original network. The general weighted pruning algorithm can only reduce the number of parameters based on the original network structure, but cannot reduce the width and depth of the pruned network. While the knowledge distillation algorithm can solve the problem by compressing the network structure, it cannot make further modifications on the processed network. To further reduce the network structure, we propose a model compression algorithm, HKDP, a hybrid method combining knowledge distillation and network pruning that can significantly reduce the overall size of the network and maintain substantial accuracy. This approach obtains the advantages of knowledge distillation and pruning, which achieves 10 times higher compression rate and 2 percent higher accuracy than using either algorithm alone. Concretely, we apply a stage-wise knowledge distillation algorithm in the front that can quickly and efficiently reduce the original model structure; we also apply a Stochastic Gradient Descent (SGD) based pruning method and introduce the concept of global sparsity, which allows us to customize the compression rate of the model. Our experiments on CIFAR-10 and MNIST show that our hybrid optimization algorithm has higher model accuracy and model compression ratio compared to other competitors' network compression algorithms.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123197528","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 : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674092
Collins Sey, Hang Lei, Weizhong Qian, Xiaoyu Li, Linda Delali Fiasam, Ruchao Sha, Zirui He
In recent years, the smart city paradigm continues to receive major advancements which is helping to improve the quality of life of people within the environment. The Internet of Things (IoT) which represents the backbone of the Smart City paradigm is receiving exponential growth. This exponential growth is also companied by some challenge which need to be addressed to further support the ever-growing demand of the IoT devices. Secure firmware update and distribution mechanisms is a major stage in the lifecycle of IoT devices management. Although the Internet Engineering Task Force (IETF) Software Updates for Internet of Things (SUIT) have started preparing software update models for IoT devices, scalability of secure firmware update distribution and centralization exists as challenges for the current model. In this paper, we propose a blockchain based firmware update architecture for IoT devices. The proposed architecture ensures secure distribution of firmware updates, malware-proof and solves the author-disappearing issue. We introduced a key revocation mechanism to secure the IoT environment from malicious devices. We further secure centralized entities that are susceptible to targeting attacks and single point of failure problem that is critical to the system by integrating all activities into the blockchain as transactions. The proposed model in this paper achieves effective and efficient security for IoT device update, as well as addressing the targeting attack and the author-disappearing issue in IoT device management.
{"title":"Firmblock: A Scalable Blockchain-Based Malware-Proof Firmware Update Architecture with Revocation for IoT Devices","authors":"Collins Sey, Hang Lei, Weizhong Qian, Xiaoyu Li, Linda Delali Fiasam, Ruchao Sha, Zirui He","doi":"10.1109/ICCWAMTIP53232.2021.9674092","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674092","url":null,"abstract":"In recent years, the smart city paradigm continues to receive major advancements which is helping to improve the quality of life of people within the environment. The Internet of Things (IoT) which represents the backbone of the Smart City paradigm is receiving exponential growth. This exponential growth is also companied by some challenge which need to be addressed to further support the ever-growing demand of the IoT devices. Secure firmware update and distribution mechanisms is a major stage in the lifecycle of IoT devices management. Although the Internet Engineering Task Force (IETF) Software Updates for Internet of Things (SUIT) have started preparing software update models for IoT devices, scalability of secure firmware update distribution and centralization exists as challenges for the current model. In this paper, we propose a blockchain based firmware update architecture for IoT devices. The proposed architecture ensures secure distribution of firmware updates, malware-proof and solves the author-disappearing issue. We introduced a key revocation mechanism to secure the IoT environment from malicious devices. We further secure centralized entities that are susceptible to targeting attacks and single point of failure problem that is critical to the system by integrating all activities into the blockchain as transactions. The proposed model in this paper achieves effective and efficient security for IoT device update, as well as addressing the targeting attack and the author-disappearing issue in IoT device management.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125144103","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 : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674056
Wang Juan, C. Shimin, Zhao Jun, Han Bin, Shi Lei
In order to identify Tor anonymous network traffic which was generated by the most widely used anonymous network in the world, analyzing the features that can be used to recognize Tor traffic based on meek pluggable transport, and proposing a method based on machine learning to classify Tor traffic. Tor traffic identification aimed at Tor-Meek traffic which using Meek traffic confusion technology in Tor network. To determine the flow characteristics that can be used to identify Tor traffic from the original feature set, RandomForest feature selection method is used to evaluate the importance of these features, and select the available feature subset. The Tor traffic classifier is constructed by using C4.5, RandomForest and KNN algorithms to identify Tor traffic. Experiment shows that Tor traffic identification methods based on three classification algorithms can effectively identify Tor anonymous network traffic, for different versions of Tor client, the precise and recall are all greater than 94% when identify Tor traffic.
{"title":"Identification of Tor Anonymous Network Traffic Based on Machine Learning","authors":"Wang Juan, C. Shimin, Zhao Jun, Han Bin, Shi Lei","doi":"10.1109/ICCWAMTIP53232.2021.9674056","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674056","url":null,"abstract":"In order to identify Tor anonymous network traffic which was generated by the most widely used anonymous network in the world, analyzing the features that can be used to recognize Tor traffic based on meek pluggable transport, and proposing a method based on machine learning to classify Tor traffic. Tor traffic identification aimed at Tor-Meek traffic which using Meek traffic confusion technology in Tor network. To determine the flow characteristics that can be used to identify Tor traffic from the original feature set, RandomForest feature selection method is used to evaluate the importance of these features, and select the available feature subset. The Tor traffic classifier is constructed by using C4.5, RandomForest and KNN algorithms to identify Tor traffic. Experiment shows that Tor traffic identification methods based on three classification algorithms can effectively identify Tor anonymous network traffic, for different versions of Tor client, the precise and recall are all greater than 94% when identify Tor traffic.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121200941","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：