In this study,we introduce our newly developed measurement-fed-perception self-adaption Low-cost UAV Coordinated Carbon Observation Network(LUCCN)prototype.The LUCCN primarily consists of two categories of instruments...In this study,we introduce our newly developed measurement-fed-perception self-adaption Low-cost UAV Coordinated Carbon Observation Network(LUCCN)prototype.The LUCCN primarily consists of two categories of instruments,including ground-based and UAV-based in-situ measurement.We use the GMP343,a low-cost non-dispersive infrared sensor,in both ground-based and UAV-based instruments.The first integrated measurement campaign took place in Shenzhen,China,4 May 2023.During the campaign,we found that LUCCN’s UAV component presented significant data-collecting advantages over its ground-based counterpart owing to the relatively high altitudes of the point emission sources,which was especially obvious at a gas power plant in Shenzhen.The emission flux was calculated by a crosssectional flux(CSF)method,the results of which differed from the Open-Data Inventory for Anthropogenic Carbon dioxide(ODIAC).The CSF result was slightly larger than others because of the low sampling rate of the whole emission cross section.The LUCCN system will be applied in future carbon monitoring campaigns to increase the spatiotemporal coverage of carbon emission information,especially in scenarios involving the detection of smaller-scale,rapidly varying sources and sinks.展开更多
The Distributed Network Performance Measurement Sys-tem provides functions to derive performance indices of networks and services, which are significant for Network Management System. To make these two systems coopera...The Distributed Network Performance Measurement Sys-tem provides functions to derive performance indices of networks and services, which are significant for Network Management System. To make these two systems cooperate, we realize this cross-system invocation platform, using Web Service, a mechanism which allows two systems to exchange data over the internet through publishing interfaces [1]. There are several mature Web Service frameworks, Apache Axis2, Apache CXF etc. In this paper we choose Apache Axis2 to achieve the objective that the Network Management System can invocate the net-work performance measurement functions via the Web Services.展开更多
With the advent of large-scale and high-speed IPv6 network technology, an effective multi-point traffic sampling is becoming a necessity. A distributed multi-point traffic sampling method that provides an accurate and...With the advent of large-scale and high-speed IPv6 network technology, an effective multi-point traffic sampling is becoming a necessity. A distributed multi-point traffic sampling method that provides an accurate and efficient solution to measure IPv6 traffic is proposed. The proposed method is to sample IPv6 traffic based on the analysis of bit randomness of each byte in the packet header. It offers a way to consistently select the same subset of packets at each measurement point, which satisfies the requirement of the distributed multi-point measurement. Finally, using real IPv6 traffic traces, the conclusion that the sampled traffic data have a good uniformity that satisfies the requirement of sampling randomness and can correctly reflect the packet size distribution of full packet trace is proved.展开更多
With the rapid growth of network bandwidth,traffic identification is currently an important challenge for network management and security.In recent years,packet sampling has been widely used in most network management...With the rapid growth of network bandwidth,traffic identification is currently an important challenge for network management and security.In recent years,packet sampling has been widely used in most network management systems.In this paper,in order to improve the accuracy of network traffic identification,sampled NetFlow data is applied to traffic identification,and the impact of packet sampling on the accuracy of the identification method is studied.This study includes feature selection,a metric correlation analysis for the application behavior,and a traffic identification algorithm.Theoretical analysis and experimental results show that the significance of behavior characteristics becomes lower in the packet sampling environment.Meanwhile,in this paper,the correlation analysis results in different trends according to different features.However,as long as the flow number meets the statistical requirement,the feature selection and the correlation degree will be independent of the sampling ratio.While in a high sampling ratio,where the effective information would be less,the identification accuracy is much lower than the unsampled packets.Finally,in order to improve the accuracy of the identification,we propose a Deep Belief Networks Application Identification(DBNAI)method,which can achieve better classification performance than other state-of-the-art methods.展开更多
Influences of the clock resolution of bandwidth estimator on the accuracy and stability of the packet pair algorithm was analyzed.A mathematic model has been established to reveal the relationship between the result d...Influences of the clock resolution of bandwidth estimator on the accuracy and stability of the packet pair algorithm was analyzed.A mathematic model has been established to reveal the relationship between the result deviation coefficient and the packet size,clock resolution and real bandwidth(value)of the measured route.A bandwidth self-adapting packet pair algorithm was presented based on the mathematic model to reduce the estimation error resulting from the clock resolution and to improve the accuracy and stability of measurement by adjusting the deviation coefficient.Experimental results have verified the validity and stability of the algorithm.展开更多
The main objective of the present study is the development of a new algorithm that can adapt to complex and changeable environments.An artificial fish swarm algorithm is developed which relies on a wireless sensor net...The main objective of the present study is the development of a new algorithm that can adapt to complex and changeable environments.An artificial fish swarm algorithm is developed which relies on a wireless sensor network(WSN)in a hydrodynamic background.The nodes of this algorithm are viscous fluids and artificial fish,while related‘events’are directly connected to the food available in the related virtual environment.The results show that the total processing time of the data by the source node is 6.661 ms,of which the processing time of crosstalk data is 3.789 ms,accounting for 56.89%.The total processing time of the data by the relay node is 15.492 ms,of which the system scheduling and the Carrier Sense Multiple Access(CSMA)rollback time of the forwarding is 8.922 ms,accounting for 57.59%.The total time for the data processing of the receiving node is 11.835 ms,of which the processing time of crosstalk data is 3.791 ms,accounting for 32.02%;the serial data processing time is 4.542 ms,accounting for 38.36%.Crosstalk packets occupy a certain amount of system overhead in the internal communication of nodes,which is one of the causes of node-level congestion.We show that optimizing the crosstalk phenomenon can alleviate the internal congestion of nodes to some extent.展开更多
The revolution of information technology within the p ast twenty years has dramatically changed the picture of our economy. Numerous n ew possibilities of communication have let competition advantages for many compa n...The revolution of information technology within the p ast twenty years has dramatically changed the picture of our economy. Numerous n ew possibilities of communication have let competition advantages for many compa nies and even advantageous macroeconomic consequences emerge on national and international level. Through newly developed information technologies the knowl edge base of market participants improves with a concurrent reduction of the inf ormation obtaining costs. As a result considerable competition advantages develo p for those companies acting in E-commerce networks. These advantages of the la test development lead to macroeconomic effects on national level, if the effecti veness and efficiency increasing possibilities are used more strongly than in ot her countries. Positive international effects arise since the allocation effic iency is increased through intensified competition between different market pa rticipants in various countries. This in turn leads to an increase in wo rldwide prosperity. This causal chain however is not yet realistic to the whole extent, as such an i ncreased transparency of information is not necessarily accepted by all market p articipants. Otherwise a considerable productivity increase would already have o ccurred in industrial countries. Overall the question arises, whether the change s in the competition situation make single enterprises technically more effectiv e, concurrently however deteriorate the efficiency of the entire market through informational asymmetries. To answer these and further questions and to measure the effectiveness and effic iency of various E-commerce networks an interdisciplinary analysis platform is to be developed. With the help of this platform, it should be possible to examin e single and macroeconomic questions, reveal temporal connections and to analyse aspects of business management and national economy, information management, em ployment politics and finance politics. For this, various part-models for the i ndividual knowledge disciplines have to be generated and brought together in the platform. This platform allows various users to make the right decisions (effec tiveness) with the help of the developed models and to competently estimate the effects (efficiency). Currently models of the individual knowledge disciplines (business management, e conomics, computer science) are being developed within the research project EEE. con. This project deals with the question of Supply Chain Management (SCM), E-P rocurement, with the implementation of inter-organisational information systems , as well as various market, competition and organisation models. The department of economics and computer science from Prof. Dr.-Ing. habil. W. Dangelmaier particularly deals with the development of an agent controlled SCM- communication model which is part of the E-commerce analysis platform. Both are described in this paper. Furthermore, a unified modelling language in order to allow a prototypic implementation of the analysis tool and to make the work with other project participants and external participants easier is decided upon wit hin this project.展开更多
A network perspective has increasingly become an organizational paradigm for understanding regional spatial structures. Based on a critical overview of existing empirical models for estimating intercity networks based...A network perspective has increasingly become an organizational paradigm for understanding regional spatial structures. Based on a critical overview of existing empirical models for estimating intercity networks based on firm linkages, this study extends the re- cently proposed regional corporate city model algorithm by proposing a new method for ap- proximating urban networks based on the Iocational strategies of firms. The new method considers both regional and hierarchical network features and avoids the information loss associated with the conversion from two-mode firm-city networks to one-mode city-city networks. In addition, networks estimated by using the method proposed herein are suitable when employing social network analysis. Finally, this method is empirically validated by ex- amining intercity firm networks formed by advanced producer services firms in China's two largest metropolitan areas, namely the Yangtze River Delta and Pearl River Delta. The pre- sented empirical analysis suggests two main findings. First, in contrast to conventional methods (e.g., the interlocking city network model), our new method produces regional and hierarchical urban networks that more closely resemble reality. Second, the new method al- lows us to use social network analysis to assess betweenness and closeness centralities. However, regardless of the model applied, the validity of any method that measures urban networks depends on the soundness of its underlying assumptions about how network actors (firms, in our case) interact.展开更多
Classification of network traffic is the essential step for many network researches. However, with the rapid evolution of Internet applications the effectiveness of the port-based or payload-based identification appro...Classification of network traffic is the essential step for many network researches. However, with the rapid evolution of Internet applications the effectiveness of the port-based or payload-based identification approaches has been greatly diminished in recent years. And many researchers begin to turn their attentions to an alternative machine learning based method. This paper presents a novel machine learning-based classification model, which combines ensemble learning paradigm with co-training techniques. Compared to previous approaches, most of which only employed single classifier, multiple classifters and semi-supervised learning are applied in our method and it mainly helps to overcome three shortcomings: limited flow accuracy rate, weak adaptability and huge demand of labeled training set. In this paper, statistical characteristics of IP flows are extracted from the packet level traces to establish the feature set, then the classification model is crested and tested and the empirical results prove its feasibility and effectiveness.展开更多
Increased adoption of smartphones leads to the explosive growth of mobile network traffic. Understanding the traffic characteristics of mobile network is important for Intemet service providers (ISPs) to optimize ne...Increased adoption of smartphones leads to the explosive growth of mobile network traffic. Understanding the traffic characteristics of mobile network is important for Intemet service providers (ISPs) to optimize network resources. In this paper, we conduct a detailed measurement study on the hyper text transfer protocol (HTTP) traffic characteristics of cellular network among different operating systems (OSs) as well as different device-types. Firstly, we propose a probability-based method to identify the installed OS of eacb smartphone. Then we analyze the traffic characteristics of these smartphones in terms of OS and device-type based on a dataset across 31 days (a billing cycle). Finally, we identify the installed apps of each smartphone and compare the usage of apps on the dimensions of OS and device-type. Our measurement study provides insights for network operators to strategize priciag and resource allocation for their cellular data networks.展开更多
In this paper a feedback control mechanism of Quality of Service (QoS) management is proposed. By measuring and monitoring the operational information of the IP telecommunication network, the feedback control-loops ...In this paper a feedback control mechanism of Quality of Service (QoS) management is proposed. By measuring and monitoring the operational information of the IP telecommunication network, the feedback control-loops tune the network according to the trends of the traffic characteristics in order to gain a favorable network status. With theoretical analysis, it is concluded that the QoS o fan arbitrary k-class flow can be guaranteed when the capacity of the k-class traffic is not fully filled. Thus by distributing the bandwidth of the egress bottleneck link among the ingress links, the congestion can be relieved. Moreover, the capacity of each traffic class that shares the same link affects one another. Therefore, the feedback control also aims at adjusting the capacity proportion among the traffic classes. In order to realize the goals, three feedback control-loops are designed in the QoS management system. A CORBA-based implementation is deployed on the testbed .展开更多
Location based services(LBS)are widely utilized,and determining the location of users’IP is the foundation for LBS.Constrained by unstable delay and insufficient landmarks,the existing geolocation algorithms have pro...Location based services(LBS)are widely utilized,and determining the location of users’IP is the foundation for LBS.Constrained by unstable delay and insufficient landmarks,the existing geolocation algorithms have problems such as low geolocation accuracy and uncertain geolocation error,difficult to meet the requirements of LBS for accuracy and reliability.A new IP geolocation algorithm based on router error training is proposed in this manuscript to improve the accuracy of geolocation results and obtain the current geolocation error range.Firstly,bootstrapping is utilized to divide the landmark data into training set and verification set,and/24 subnet distribution is utilized to extend the training set.Secondly,the path detection is performed on nodes in the three data sets respectively to extract the metropolitan area network(MAN)of the target city,and the geolocation result and error of each router in MAN are obtained by training the detection results.Finally,the MAN is utilized to get the target’s location.Based on China’s 24,254 IP geolocation experiments,the proposed algorithm has higher geolocation accuracy and lower median error than existing typical geolocation algorithms LBG,SLG,NNG and RNBG,and in most cases the difference is less than 10km between estimated error and actual error.展开更多
Although many classical IP geolocation algorithms are suitable to rich-connected networks, their performances are seriously affected in poor-connected networks with weak delay-distance correlation. This paper tries to...Although many classical IP geolocation algorithms are suitable to rich-connected networks, their performances are seriously affected in poor-connected networks with weak delay-distance correlation. This paper tries to improve the performances of classical IP geolocation algorithms by finding rich-connected sub-networks inside poor-connected networks. First, a new delay-distance correlation model (RTD-Corr model) is proposed. It builds the relationship between delay-distance correlation and actual network factors such as the tortuosity of the network path and the ratio of propagation delay. Second, based on the RTD-Corr model and actual network characteristics, this paper discusses about how to find rich-connected networks inside China Intemet which is a typical actual poor-connected network. Then we find rich-connected sub-networks of China Intemet through a large-scale network measurement which covers three major ISPs and thirty provinces. At last, based on the founded rich-connected sub-networks, we modify two classical IP geolocation algorithms and the experiments in China Intemet show that their accuracy is significantly increased.展开更多
Sybil attacks are one kind of well-known and powerful attacks against online social networks (OSNs). In a sybil attack, a malicious attacker generates a sybil group consisting of multiple sybil users, and controls t...Sybil attacks are one kind of well-known and powerful attacks against online social networks (OSNs). In a sybil attack, a malicious attacker generates a sybil group consisting of multiple sybil users, and controls them to attack the system. However, data confidentiality policies of major social network providers have severely limited researchers' access to large-scale datasets of sybil groups. A deep understanding of sybil groups can provide important insights into the characteristics of malicious behavior, as well as numerous practical implications on the design of security mechanisms. In this paper, we present an initial study to measure sybil groups in a large-scale OSN, Renren. We analyze sybil groups at different levels, including individual information, social relationships, and malicious activities. Our main observations are: 1) user information in sybil groups is usually incomplete and in poor quality; 2) sybil groups have special evolution patterns in connectivity structure, including bursty actions to add nodes, and a monotonous merging pattern that lacks non-singleton mergings; 3) several sybil groups have strong relationships with each other and compose sybil communities, and these communities cover a large number of users and pose great potential threats; 4) some sybil users are not banned until a long time after registration in some sybil groups. The characteristics of sybil groups can be leveraged to improve the security mechanisms in OSNs to defend against sybil attacks. Specifically, we suggest that OSNs should 1) check information completeness and quality, 2) learn from dynamics of community connectivity structure to detect sybil groups, 3) monitor sybil communities and inspect them carefully to prevent collusion, and 4) inspect sybil groups that behave normally even for a long time to prevent potential malicious behaviors.展开更多
基金supported by the National Key Research and Development Plan(Grant No.2021YFB3901000)the Chinese Academy of Sciences Project for Young Scientists in Basic Research(YSBR-037)+2 种基金the International Partnership Program of the Chinese Academy of Sciences(060GJHZ2022070MI)the MOST-ESA Dragon-5 Programme for Monitoring Greenhouse Gases from Space(ID.59355)the Finland–China Mobility Cooperation Project funded by the Academy of Finland(No.348596)。
文摘In this study,we introduce our newly developed measurement-fed-perception self-adaption Low-cost UAV Coordinated Carbon Observation Network(LUCCN)prototype.The LUCCN primarily consists of two categories of instruments,including ground-based and UAV-based in-situ measurement.We use the GMP343,a low-cost non-dispersive infrared sensor,in both ground-based and UAV-based instruments.The first integrated measurement campaign took place in Shenzhen,China,4 May 2023.During the campaign,we found that LUCCN’s UAV component presented significant data-collecting advantages over its ground-based counterpart owing to the relatively high altitudes of the point emission sources,which was especially obvious at a gas power plant in Shenzhen.The emission flux was calculated by a crosssectional flux(CSF)method,the results of which differed from the Open-Data Inventory for Anthropogenic Carbon dioxide(ODIAC).The CSF result was slightly larger than others because of the low sampling rate of the whole emission cross section.The LUCCN system will be applied in future carbon monitoring campaigns to increase the spatiotemporal coverage of carbon emission information,especially in scenarios involving the detection of smaller-scale,rapidly varying sources and sinks.
文摘The Distributed Network Performance Measurement Sys-tem provides functions to derive performance indices of networks and services, which are significant for Network Management System. To make these two systems cooperate, we realize this cross-system invocation platform, using Web Service, a mechanism which allows two systems to exchange data over the internet through publishing interfaces [1]. There are several mature Web Service frameworks, Apache Axis2, Apache CXF etc. In this paper we choose Apache Axis2 to achieve the objective that the Network Management System can invocate the net-work performance measurement functions via the Web Services.
基金This project was supported by the National Natural Science Foundation of China (60572147,60132030)
文摘With the advent of large-scale and high-speed IPv6 network technology, an effective multi-point traffic sampling is becoming a necessity. A distributed multi-point traffic sampling method that provides an accurate and efficient solution to measure IPv6 traffic is proposed. The proposed method is to sample IPv6 traffic based on the analysis of bit randomness of each byte in the packet header. It offers a way to consistently select the same subset of packets at each measurement point, which satisfies the requirement of the distributed multi-point measurement. Finally, using real IPv6 traffic traces, the conclusion that the sampled traffic data have a good uniformity that satisfies the requirement of sampling randomness and can correctly reflect the packet size distribution of full packet trace is proved.
基金supported by Key Scientific and Technological Research Projects in Henan Province(Grand No 192102210125)Key scientific research projects of colleges and universities in Henan Province(23A520054)Open Foundation of State key Laboratory of Networking and Switching Technology(Beijing University of Posts and Telecommunications)(SKLNST-2020-2-01).
文摘With the rapid growth of network bandwidth,traffic identification is currently an important challenge for network management and security.In recent years,packet sampling has been widely used in most network management systems.In this paper,in order to improve the accuracy of network traffic identification,sampled NetFlow data is applied to traffic identification,and the impact of packet sampling on the accuracy of the identification method is studied.This study includes feature selection,a metric correlation analysis for the application behavior,and a traffic identification algorithm.Theoretical analysis and experimental results show that the significance of behavior characteristics becomes lower in the packet sampling environment.Meanwhile,in this paper,the correlation analysis results in different trends according to different features.However,as long as the flow number meets the statistical requirement,the feature selection and the correlation degree will be independent of the sampling ratio.While in a high sampling ratio,where the effective information would be less,the identification accuracy is much lower than the unsampled packets.Finally,in order to improve the accuracy of the identification,we propose a Deep Belief Networks Application Identification(DBNAI)method,which can achieve better classification performance than other state-of-the-art methods.
基金This workis supported by973Project(National Keystone Foundation Research Project,No.G199903271)the National Natural Science Foundation of China(No.90104022)the National High Technology Development Program of China(No.2001AA112120,No.2002AA104550).
文摘Influences of the clock resolution of bandwidth estimator on the accuracy and stability of the packet pair algorithm was analyzed.A mathematic model has been established to reveal the relationship between the result deviation coefficient and the packet size,clock resolution and real bandwidth(value)of the measured route.A bandwidth self-adapting packet pair algorithm was presented based on the mathematic model to reduce the estimation error resulting from the clock resolution and to improve the accuracy and stability of measurement by adjusting the deviation coefficient.Experimental results have verified the validity and stability of the algorithm.
基金financially supported by Natural Science Foundation of Heilongjiang Province of China[Grant No.LH2019F042].
文摘The main objective of the present study is the development of a new algorithm that can adapt to complex and changeable environments.An artificial fish swarm algorithm is developed which relies on a wireless sensor network(WSN)in a hydrodynamic background.The nodes of this algorithm are viscous fluids and artificial fish,while related‘events’are directly connected to the food available in the related virtual environment.The results show that the total processing time of the data by the source node is 6.661 ms,of which the processing time of crosstalk data is 3.789 ms,accounting for 56.89%.The total processing time of the data by the relay node is 15.492 ms,of which the system scheduling and the Carrier Sense Multiple Access(CSMA)rollback time of the forwarding is 8.922 ms,accounting for 57.59%.The total time for the data processing of the receiving node is 11.835 ms,of which the processing time of crosstalk data is 3.791 ms,accounting for 32.02%;the serial data processing time is 4.542 ms,accounting for 38.36%.Crosstalk packets occupy a certain amount of system overhead in the internal communication of nodes,which is one of the causes of node-level congestion.We show that optimizing the crosstalk phenomenon can alleviate the internal congestion of nodes to some extent.
文摘The revolution of information technology within the p ast twenty years has dramatically changed the picture of our economy. Numerous n ew possibilities of communication have let competition advantages for many compa nies and even advantageous macroeconomic consequences emerge on national and international level. Through newly developed information technologies the knowl edge base of market participants improves with a concurrent reduction of the inf ormation obtaining costs. As a result considerable competition advantages develo p for those companies acting in E-commerce networks. These advantages of the la test development lead to macroeconomic effects on national level, if the effecti veness and efficiency increasing possibilities are used more strongly than in ot her countries. Positive international effects arise since the allocation effic iency is increased through intensified competition between different market pa rticipants in various countries. This in turn leads to an increase in wo rldwide prosperity. This causal chain however is not yet realistic to the whole extent, as such an i ncreased transparency of information is not necessarily accepted by all market p articipants. Otherwise a considerable productivity increase would already have o ccurred in industrial countries. Overall the question arises, whether the change s in the competition situation make single enterprises technically more effectiv e, concurrently however deteriorate the efficiency of the entire market through informational asymmetries. To answer these and further questions and to measure the effectiveness and effic iency of various E-commerce networks an interdisciplinary analysis platform is to be developed. With the help of this platform, it should be possible to examin e single and macroeconomic questions, reveal temporal connections and to analyse aspects of business management and national economy, information management, em ployment politics and finance politics. For this, various part-models for the i ndividual knowledge disciplines have to be generated and brought together in the platform. This platform allows various users to make the right decisions (effec tiveness) with the help of the developed models and to competently estimate the effects (efficiency). Currently models of the individual knowledge disciplines (business management, e conomics, computer science) are being developed within the research project EEE. con. This project deals with the question of Supply Chain Management (SCM), E-P rocurement, with the implementation of inter-organisational information systems , as well as various market, competition and organisation models. The department of economics and computer science from Prof. Dr.-Ing. habil. W. Dangelmaier particularly deals with the development of an agent controlled SCM- communication model which is part of the E-commerce analysis platform. Both are described in this paper. Furthermore, a unified modelling language in order to allow a prototypic implementation of the analysis tool and to make the work with other project participants and external participants easier is decided upon wit hin this project.
基金National Natural Science Foundation of China, No.51478189 No.41401178+2 种基金 National Social Science Fund, 1 l&ZD154 State Key Laboratory of Subtropical Building Science, No.2013KB20 Fundamental Research Funds for the Central Universities, No.2013ZZ0022
文摘A network perspective has increasingly become an organizational paradigm for understanding regional spatial structures. Based on a critical overview of existing empirical models for estimating intercity networks based on firm linkages, this study extends the re- cently proposed regional corporate city model algorithm by proposing a new method for ap- proximating urban networks based on the Iocational strategies of firms. The new method considers both regional and hierarchical network features and avoids the information loss associated with the conversion from two-mode firm-city networks to one-mode city-city networks. In addition, networks estimated by using the method proposed herein are suitable when employing social network analysis. Finally, this method is empirically validated by ex- amining intercity firm networks formed by advanced producer services firms in China's two largest metropolitan areas, namely the Yangtze River Delta and Pearl River Delta. The pre- sented empirical analysis suggests two main findings. First, in contrast to conventional methods (e.g., the interlocking city network model), our new method produces regional and hierarchical urban networks that more closely resemble reality. Second, the new method al- lows us to use social network analysis to assess betweenness and closeness centralities. However, regardless of the model applied, the validity of any method that measures urban networks depends on the soundness of its underlying assumptions about how network actors (firms, in our case) interact.
基金Supported by the National Natural Science Foundation of China (Grant Nos.60525213 and 60776096)the National Basic Research Program of China (Grant No.2006CB303106)+2 种基金the National High-Tech Research & Development Program of China (Grant Nos.2007AA01Z236 and 2007AA01Z449)the Joint Funds of NSFC-Guangdong (Grant No.U0735001)the National Project of Scientific and Technical Supporting Programs (Grant No.2007BAH13B01)
文摘Classification of network traffic is the essential step for many network researches. However, with the rapid evolution of Internet applications the effectiveness of the port-based or payload-based identification approaches has been greatly diminished in recent years. And many researchers begin to turn their attentions to an alternative machine learning based method. This paper presents a novel machine learning-based classification model, which combines ensemble learning paradigm with co-training techniques. Compared to previous approaches, most of which only employed single classifier, multiple classifters and semi-supervised learning are applied in our method and it mainly helps to overcome three shortcomings: limited flow accuracy rate, weak adaptability and huge demand of labeled training set. In this paper, statistical characteristics of IP flows are extracted from the packet level traces to establish the feature set, then the classification model is crested and tested and the empirical results prove its feasibility and effectiveness.
文摘Increased adoption of smartphones leads to the explosive growth of mobile network traffic. Understanding the traffic characteristics of mobile network is important for Intemet service providers (ISPs) to optimize network resources. In this paper, we conduct a detailed measurement study on the hyper text transfer protocol (HTTP) traffic characteristics of cellular network among different operating systems (OSs) as well as different device-types. Firstly, we propose a probability-based method to identify the installed OS of eacb smartphone. Then we analyze the traffic characteristics of these smartphones in terms of OS and device-type based on a dataset across 31 days (a billing cycle). Finally, we identify the installed apps of each smartphone and compare the usage of apps on the dimensions of OS and device-type. Our measurement study provides insights for network operators to strategize priciag and resource allocation for their cellular data networks.
基金This work is supported by National Natural Science Foundation of China (90204003) , National "863" High Technology Projects of China(2003AA121220 ,2002AA103063) , RFDR(20010013003) and Excellent Young Teachers Programof MOE,China.
文摘In this paper a feedback control mechanism of Quality of Service (QoS) management is proposed. By measuring and monitoring the operational information of the IP telecommunication network, the feedback control-loops tune the network according to the trends of the traffic characteristics in order to gain a favorable network status. With theoretical analysis, it is concluded that the QoS o fan arbitrary k-class flow can be guaranteed when the capacity of the k-class traffic is not fully filled. Thus by distributing the bandwidth of the egress bottleneck link among the ingress links, the congestion can be relieved. Moreover, the capacity of each traffic class that shares the same link affects one another. Therefore, the feedback control also aims at adjusting the capacity proportion among the traffic classes. In order to realize the goals, three feedback control-loops are designed in the QoS management system. A CORBA-based implementation is deployed on the testbed .
基金This work was supported by the National Natural Science Foundation of China(Grant Nos.U1804263,U1636219)the Science and Technology Innovation Talent Project of Henan Province(184200510018)Zhongyuan Science and Technology Innovation Leading Talent Project(214200510019).
文摘Location based services(LBS)are widely utilized,and determining the location of users’IP is the foundation for LBS.Constrained by unstable delay and insufficient landmarks,the existing geolocation algorithms have problems such as low geolocation accuracy and uncertain geolocation error,difficult to meet the requirements of LBS for accuracy and reliability.A new IP geolocation algorithm based on router error training is proposed in this manuscript to improve the accuracy of geolocation results and obtain the current geolocation error range.Firstly,bootstrapping is utilized to divide the landmark data into training set and verification set,and/24 subnet distribution is utilized to extend the training set.Secondly,the path detection is performed on nodes in the three data sets respectively to extract the metropolitan area network(MAN)of the target city,and the geolocation result and error of each router in MAN are obtained by training the detection results.Finally,the MAN is utilized to get the target’s location.Based on China’s 24,254 IP geolocation experiments,the proposed algorithm has higher geolocation accuracy and lower median error than existing typical geolocation algorithms LBG,SLG,NNG and RNBG,and in most cases the difference is less than 10km between estimated error and actual error.
基金Supported by the National Natural Science Foundation of China(61379151,61274189,61302159 and 61401512)the Excellent Youth Foundation of Henan Province of China(144100510001)Foundation of Science and Technology on Information Assurance Laboratory(KJ-14-108)
文摘Although many classical IP geolocation algorithms are suitable to rich-connected networks, their performances are seriously affected in poor-connected networks with weak delay-distance correlation. This paper tries to improve the performances of classical IP geolocation algorithms by finding rich-connected sub-networks inside poor-connected networks. First, a new delay-distance correlation model (RTD-Corr model) is proposed. It builds the relationship between delay-distance correlation and actual network factors such as the tortuosity of the network path and the ratio of propagation delay. Second, based on the RTD-Corr model and actual network characteristics, this paper discusses about how to find rich-connected networks inside China Intemet which is a typical actual poor-connected network. Then we find rich-connected sub-networks of China Intemet through a large-scale network measurement which covers three major ISPs and thirty provinces. At last, based on the founded rich-connected sub-networks, we modify two classical IP geolocation algorithms and the experiments in China Intemet show that their accuracy is significantly increased.
文摘Sybil attacks are one kind of well-known and powerful attacks against online social networks (OSNs). In a sybil attack, a malicious attacker generates a sybil group consisting of multiple sybil users, and controls them to attack the system. However, data confidentiality policies of major social network providers have severely limited researchers' access to large-scale datasets of sybil groups. A deep understanding of sybil groups can provide important insights into the characteristics of malicious behavior, as well as numerous practical implications on the design of security mechanisms. In this paper, we present an initial study to measure sybil groups in a large-scale OSN, Renren. We analyze sybil groups at different levels, including individual information, social relationships, and malicious activities. Our main observations are: 1) user information in sybil groups is usually incomplete and in poor quality; 2) sybil groups have special evolution patterns in connectivity structure, including bursty actions to add nodes, and a monotonous merging pattern that lacks non-singleton mergings; 3) several sybil groups have strong relationships with each other and compose sybil communities, and these communities cover a large number of users and pose great potential threats; 4) some sybil users are not banned until a long time after registration in some sybil groups. The characteristics of sybil groups can be leveraged to improve the security mechanisms in OSNs to defend against sybil attacks. Specifically, we suggest that OSNs should 1) check information completeness and quality, 2) learn from dynamics of community connectivity structure to detect sybil groups, 3) monitor sybil communities and inspect them carefully to prevent collusion, and 4) inspect sybil groups that behave normally even for a long time to prevent potential malicious behaviors.