In this paper, the authors investigate the overload problem of a congested router in TCP （Transmission Control Protocol） networks, with a design of an AQM （Active Queue Management） controller based on control theo...In this paper, the authors investigate the overload problem of a congested router in TCP （Transmission Control Protocol） networks, with a design of an AQM （Active Queue Management） controller based on control theory. The model of TCP/AQM is converted into a second-order system with time delay, the stability of the closed-loop AQM system is described in terms of a characteristic quasi-polynomial. The authors focus on the synthesis of first-order AQM controller. The authors＇ approach consists of determining the stabilizing regions in the parameter space of the first-order controller, then choosing controller＇s parameters within these regions. Finally, the authors illustrate the proposed methodology with an example and simulations using NS-2 simulator.展开更多
An adaptive transmission control algorithm based on TCP （TCP-ATCA） is proposed to reduce the effects of long propagation delay and high link error rate of the satellite network on the performances. The flow control ...An adaptive transmission control algorithm based on TCP （TCP-ATCA） is proposed to reduce the effects of long propagation delay and high link error rate of the satellite network on the performances. The flow control and the error recovery are differentiated by combined dynamic random early detection-explicit congestion notification （DRED-ECN） algorithm, and, moreover, the pertaining congestion control methods are used in TCP-ATCA to improve the throughput. By introducing the entire recovery algorithm, the unnecessary congestion window decrease is reduced, and the throughput and fairness are improved. Simulation results show that, compared with TCP-Reno, TCP-ATCA provides a better throughput performance when the link capacity is higher （ ≥600 packet/s）, and roughly the same when it is lower. At the same time, TCP-ATCA also increases fairness and reduces transmission delay.展开更多
With the continual growth of the variety and complexity of network crime means, the traditional packet feature matching cannot detect all kinds of intrusion behaviors completely. It is urgent to reassemble network str...With the continual growth of the variety and complexity of network crime means, the traditional packet feature matching cannot detect all kinds of intrusion behaviors completely. It is urgent to reassemble network stream to perform packet processing at a semantic level above the network layer. This paper presents an efficient TCP stream reassembly mechanism for real-time processing of high-speed network traffic. By analyzing the characteristics of network stream in high-speed network and TCP connection establishment process, several polices for designing the reassembly mechanism are built. Then, the reassembly implementation is elaborated in accordance with the policies. Finally, the reassembly mechanism is compared with the traditional reassembly mechanism by the network traffic captured in a typical gigabit gateway. Experiment results illustrate that the reassembly mechanism is efficient and can satisfy the real-time property requirement of traffic analysis system in high-speed network.展开更多
The IP-based networks on aircraft serve to support Intemet services via satellites. However, in aeronautical satellite hybrid networks, the TCP protocol performance often deteriorates due to improper decreases and slo...The IP-based networks on aircraft serve to support Intemet services via satellites. However, in aeronautical satellite hybrid networks, the TCP protocol performance often deteriorates due to improper decreases and slow recovery of the congestion window. This paper proposes a window size determination and notification mechanism, onboard-gateway-based mechanism （OGBM）, which is based on the onboard gateway in the networks on aircraft. A cross-layer approach is adopted by the onboard gateway to obtain the satellite link bandwidth information. And then, by the gateway, through changing the receiver＇s advertised window field in ACK packets, TCP sources are notified of the window size of each TCP source calculated on the ground of bandwidth delay product and flow numbers. The mechanism is able to avoid improper changes of TCP window and serve multiple users. Simulation results show that the mechanism with the fairness index close to l improves TCP performance in aeronautical satellite networks.展开更多
This paper mainly introduces about how to use socket to design network program which can transfer data in different network platform in terms of TCP/IP Network Programming Standard.
Researchers have used extensive simulation and experimental studies to understand TCP performance in wireless multihop networks. In contrast, the objective of this paper is to theoretically analyze TCP performance in ...Researchers have used extensive simulation and experimental studies to understand TCP performance in wireless multihop networks. In contrast, the objective of this paper is to theoretically analyze TCP performance in this environment. By examining the case of running one TCP session over a string topology, a system model for analyzing TCP performance in multihop wireless networks is proposed, which considers packet buffering, contention of nodes for access to the wireless channel, and spatial reuse of the wireless channel. Markov chain modelling is applied to analyze this system model. Analytical results show that when the number of hops that the TCP session crosses is ?xed, the TCP throughput is independent of the TCP congestion window size. When the number of hops increases from one, the TCP throughput decreases ?rst, and then stabilizes when the number of hops becomes large. The analysis is validated by comparing the numerical and simulation results.展开更多
A new approach, named TCP-I2NC, is proposed to improve the interaction between network coding and TCP and to maximize the network utility in interference-free multi-radio multi-channel wireless mesh networks. It is gr...A new approach, named TCP-I2NC, is proposed to improve the interaction between network coding and TCP and to maximize the network utility in interference-free multi-radio multi-channel wireless mesh networks. It is grounded on a Network Utility Maxmization (NUM) formulation which can be decomposed into a rate control problem and a packet scheduling problem. The solutions to these two problems perform resource allocation among different flows. Simulations demonstrate that TCP-I2NC results in a significant throughput gain and a small delay jitter. Network resource is fairly allocated via the solution to the NUM problem and the whole system also runs stably. Moreover, TCP-I2NC is compatible with traditional TCP variants.展开更多
The conservative Additive Increase Multiplicative Decrease mechanism of traditional TCP causes the link under-utilization in the Wide Area Networks(WANs) due to the WANs' intrinsic nature of high latency and high ...The conservative Additive Increase Multiplicative Decrease mechanism of traditional TCP causes the link under-utilization in the Wide Area Networks(WANs) due to the WANs' intrinsic nature of high latency and high packet loss.To alleviate the problem,we present the design and implantation of STAG,an Acceleration Gateway with Split-TCP in the paper.STAG is built on embedded network equipment and acts as a transparent proxy.In STAG,a new improved congestion control method named Rapid TCP is adopted,which determines whether or not to decrease the congestion window based on the packet loss trend.In particular,in the fast recovery phase,it chooses different window adjustment strategies based on the current size of congestion window to achieve higher utilization.The performance validation of STAG is done on both our emulation testbed and the real wide area network.The results show that STAG with Rapid TCP effectively adapts to the high loss network environment and significantly speeds up the applications without loss of fairness.展开更多
Nowadays,high mobility scenarios have become increasingly common.The widespread adoption of High-speed Rail(HSR)in China exemplifies this trend,while more promising use cases,such as vehicle-to-everything,continue to ...Nowadays,high mobility scenarios have become increasingly common.The widespread adoption of High-speed Rail(HSR)in China exemplifies this trend,while more promising use cases,such as vehicle-to-everything,continue to emerge.However,the Internet access provided in high mobility environments stllstruggles to achieve seamless connectivity.The next generation of wireless cellular technology 5 G further poses more requirements on the endto-end evolution to fully utilize its ultra-high band-width,while existing network diagnostic tools focus on above-IP layers or below-IP layers only.We then propose HiMoDiag,which enables flexible online analysis of the network performance in a cross-layer manner,i.e.,from the top(application layer)to the bottom(physical layer).We believe HiMoDiag could greatly simplify the process of pinpointing the deficiencies of the Internet access delivery on HSR,lead to more timely optimization and ultimately help to improve the network performance.展开更多
A novel closed-loop feedback TCP/AQM(Transfer Control Protocol/Active Queue Management) model is proposed in this paper using a discrete-time Markov chain,and a way to calculate the equilibrium distribution of this mo...A novel closed-loop feedback TCP/AQM(Transfer Control Protocol/Active Queue Management) model is proposed in this paper using a discrete-time Markov chain,and a way to calculate the equilibrium distribution of this model is given.In the model,system time is divided into time slots,the bottleneck router queue model and TCP window size model in each slot are analyzed.Finally,by combining adjacent slots,an integrated TCP/AQM analytical model is developed.By this model,the average values of packets dropping ratio and queue length in the router and TCP sending rate can be estimated,hence,this model could be a useful tool for the analysis of AQMs and support the development of new AQM schemes theoretically.The proposed TCP/AQM model is extended to a TCP-UDP(Control User Datagram Protocol)/AQM model,to analyze the TCP/AQM system performance when UDP flows exist.By implementing this model on Matlab,we compare its solutions to NS2 simulated solutions,then the validity of the model to analyze the closed-loop feedback TCP/RED(Random Early Detection) system is verified.展开更多
In data centers, the transmission control protocol(TCP) incast causes catastrophic goodput degradation to applications with a many-to-one traffic pattern. In this paper, we intend to tame incast at the receiver-side a...In data centers, the transmission control protocol(TCP) incast causes catastrophic goodput degradation to applications with a many-to-one traffic pattern. In this paper, we intend to tame incast at the receiver-side application. Towards this goal, we first develop an analytical model that formulates the incast probability as a function of connection variables and network environment settings. We combine the model with the optimization theory and derive some insights into minimizing the incast probability through tuning connection variables related to applications. Then,enlightened by the analytical results, we propose an adaptive application-layer solution to the TCP incast.The solution equally allocates advertised windows to concurrent connections, and dynamically adapts the number of concurrent connections to the varying conditions. Simulation results show that our solution consistently eludes incast and achieves high goodput in various scenarios including the ones with multiple bottleneck links and background TCP traffic.展开更多
In Wireless Mesh Networks (WMNs),the performance of conventional TCP significantly deteriorates due to the unreliable wireless channel.To enhance TCP performance in WMNs,TCP/LT is proposed in this paper.It introduces ...In Wireless Mesh Networks (WMNs),the performance of conventional TCP significantly deteriorates due to the unreliable wireless channel.To enhance TCP performance in WMNs,TCP/LT is proposed in this paper.It introduces fountain codes into packet reorganization in the protocol stack of mesh gateways and mesh clients.Furthermore,it is compatible with conventional TCP.Regarded as a Performance Enhancement Proxies (PEP),a mesh gateway buffers TCP packets into several blocks.It simultaneously processes them by using fountain encoders and then sends them to mesh clients.Apart from the improvement of the throughput of a unitary TCP flow,the entire network utility maximization can also be ensured by adjusting the scale of coding blocks for each TCP flow adaptively.Simulations show that TCP/LT presents high throughput gains over single TCP in lossy links of WMNs while preserving the fairness for multiple TCPs.As losses increase,the transmission delay of TCP/LT experiences a slow linear growth in contrast to the exponential growth of TCP.展开更多
A scheme of setting a limit to the TCP sending window size is proposed to improve the TCP fairness between upload and/or download flows in wired-cum-wireless networks. The goodput and delay of the upload and download ...A scheme of setting a limit to the TCP sending window size is proposed to improve the TCP fairness between upload and/or download flows in wired-cum-wireless networks. The goodput and delay of the upload and download TCP flows arc compared to evaluate the TCP fairness for different schemes, which are the different combinations of setting a limit （64 or 4） to the sending window size and using the delayed acknowledgement （ACK） scheme or not. Extensive simulation results and analysis show that （ 1 ） for TCP download flows, setting the limit of sending window size to 4 can improve the fairness; （2） for TCP upload flows, limiting the sending window size and using the delayed ACK strategy are both beneficial to fairness; （3） for TCP download and upload mixture flows, limiting the sending window size to a small value （ e. g. , 4） rather than using the delayed ACK strategy, is the solution to improvement of the fairness ; （4） a large delay interval （200 ms or 300 ms） does not result in improvement in fairness and performance; （ 5 ） a larger TCP packet size （ 1400 B） can improve the TCP upload goodput and decrease the download goodput; in contrast, a smaller TCP packet size （560 B） can increase the download goodput and decrease the upload goodput.展开更多
Network measurement is an important approach to understand network behaviors, which has been widely studied. Both Transfer Control Protocol （TCP） and Interact Control Messages Protocol （ICMP） are applied in networ...Network measurement is an important approach to understand network behaviors, which has been widely studied. Both Transfer Control Protocol （TCP） and Interact Control Messages Protocol （ICMP） are applied in network measurement, while investigating the differences between the measured results of these two protocols is an important topic that has been less investigated. In this paper, to compare the differences between TCP and ICMP when they are used in measuring host connectivity, RTT, and packet loss rate, two groups of comparison programs have been designed, and after careful evaluation of the program parameters, a lot of comparison experiments are executed on the Internet. The experimental results show that, there are significant differences between the host connectivity measured using TCP or ICMP; in general, the accuracy of connectivity measured using TCP is 20%- 30% higher than that measured using ICMP. The case of RTT and packet loss rate is complicated, which are related to path loads and destination host loads. While commonly, the RTF and packet loss rate＂ measured using TCP or ICMP are very close. According to the experimental results, some advices are also given on protocol selection for conducting accurate connectivity, RTF and packet loss rate measurements.展开更多
基金Supported by National Natural Science Foundation of China （60974129, 70931002）, Natural Science Foundation of Jiangsu Province （BK2008188, BK2009388）, and Science Foundation of Nanjing University of Science and Technology （AB41972）
文摘In this paper, the authors investigate the overload problem of a congested router in TCP （Transmission Control Protocol） networks, with a design of an AQM （Active Queue Management） controller based on control theory. The model of TCP/AQM is converted into a second-order system with time delay, the stability of the closed-loop AQM system is described in terms of a characteristic quasi-polynomial. The authors focus on the synthesis of first-order AQM controller. The authors＇ approach consists of determining the stabilizing regions in the parameter space of the first-order controller, then choosing controller＇s parameters within these regions. Finally, the authors illustrate the proposed methodology with an example and simulations using NS-2 simulator.
基金National Natural Science Foundation of China (60502017, 60532030, 60625102)The Blue-Sky New Star Grant of Beijing University of Aeronautics and Astronautics (2004)
文摘An adaptive transmission control algorithm based on TCP （TCP-ATCA） is proposed to reduce the effects of long propagation delay and high link error rate of the satellite network on the performances. The flow control and the error recovery are differentiated by combined dynamic random early detection-explicit congestion notification （DRED-ECN） algorithm, and, moreover, the pertaining congestion control methods are used in TCP-ATCA to improve the throughput. By introducing the entire recovery algorithm, the unnecessary congestion window decrease is reduced, and the throughput and fairness are improved. Simulation results show that, compared with TCP-Reno, TCP-ATCA provides a better throughput performance when the link capacity is higher （ ≥600 packet/s）, and roughly the same when it is lower. At the same time, TCP-ATCA also increases fairness and reduces transmission delay.
基金National High-Tech Research and Development Program of China (863 Program) (No.2007AA01Z309)
文摘With the continual growth of the variety and complexity of network crime means, the traditional packet feature matching cannot detect all kinds of intrusion behaviors completely. It is urgent to reassemble network stream to perform packet processing at a semantic level above the network layer. This paper presents an efficient TCP stream reassembly mechanism for real-time processing of high-speed network traffic. By analyzing the characteristics of network stream in high-speed network and TCP connection establishment process, several polices for designing the reassembly mechanism are built. Then, the reassembly implementation is elaborated in accordance with the policies. Finally, the reassembly mechanism is compared with the traditional reassembly mechanism by the network traffic captured in a typical gigabit gateway. Experiment results illustrate that the reassembly mechanism is efficient and can satisfy the real-time property requirement of traffic analysis system in high-speed network.
基金National Natural Science Foundation of China (60532030)
文摘The IP-based networks on aircraft serve to support Intemet services via satellites. However, in aeronautical satellite hybrid networks, the TCP protocol performance often deteriorates due to improper decreases and slow recovery of the congestion window. This paper proposes a window size determination and notification mechanism, onboard-gateway-based mechanism （OGBM）, which is based on the onboard gateway in the networks on aircraft. A cross-layer approach is adopted by the onboard gateway to obtain the satellite link bandwidth information. And then, by the gateway, through changing the receiver＇s advertised window field in ACK packets, TCP sources are notified of the window size of each TCP source calculated on the ground of bandwidth delay product and flow numbers. The mechanism is able to avoid improper changes of TCP window and serve multiple users. Simulation results show that the mechanism with the fairness index close to l improves TCP performance in aeronautical satellite networks.
文摘This paper mainly introduces about how to use socket to design network program which can transfer data in different network platform in terms of TCP/IP Network Programming Standard.
文摘Researchers have used extensive simulation and experimental studies to understand TCP performance in wireless multihop networks. In contrast, the objective of this paper is to theoretically analyze TCP performance in this environment. By examining the case of running one TCP session over a string topology, a system model for analyzing TCP performance in multihop wireless networks is proposed, which considers packet buffering, contention of nodes for access to the wireless channel, and spatial reuse of the wireless channel. Markov chain modelling is applied to analyze this system model. Analytical results show that when the number of hops that the TCP session crosses is ?xed, the TCP throughput is independent of the TCP congestion window size. When the number of hops increases from one, the TCP throughput decreases ?rst, and then stabilizes when the number of hops becomes large. The analysis is validated by comparing the numerical and simulation results.
基金This work was supported by the State Key Program of Na- tional Nature Science Foundation of China under Grants No. U0835003, No. 60872087.
文摘A new approach, named TCP-I2NC, is proposed to improve the interaction between network coding and TCP and to maximize the network utility in interference-free multi-radio multi-channel wireless mesh networks. It is grounded on a Network Utility Maxmization (NUM) formulation which can be decomposed into a rate control problem and a packet scheduling problem. The solutions to these two problems perform resource allocation among different flows. Simulations demonstrate that TCP-I2NC results in a significant throughput gain and a small delay jitter. Network resource is fairly allocated via the solution to the NUM problem and the whole system also runs stably. Moreover, TCP-I2NC is compatible with traditional TCP variants.
基金supported by the National Natural Science Foundation of China(Grant nos.61173169,61103204,and 61402542)the open funding of Science and Technology on Information Transmission and Dissemination in Communication Networks Laboratory(Grant no.ITDU14010/KX142600017)the Postgraduate Innovative Research Project of Hunan Province(No.CX2012B065)
文摘The conservative Additive Increase Multiplicative Decrease mechanism of traditional TCP causes the link under-utilization in the Wide Area Networks(WANs) due to the WANs' intrinsic nature of high latency and high packet loss.To alleviate the problem,we present the design and implantation of STAG,an Acceleration Gateway with Split-TCP in the paper.STAG is built on embedded network equipment and acts as a transparent proxy.In STAG,a new improved congestion control method named Rapid TCP is adopted,which determines whether or not to decrease the congestion window based on the packet loss trend.In particular,in the fast recovery phase,it chooses different window adjustment strategies based on the current size of congestion window to achieve higher utilization.The performance validation of STAG is done on both our emulation testbed and the real wide area network.The results show that STAG with Rapid TCP effectively adapts to the high loss network environment and significantly speeds up the applications without loss of fairness.
基金supported by National Key Research and Development Plan,China(Grant No.2020YFB1710900)National Natural Science Foundation of China(Grant No.62022005 and 62172008).
文摘Nowadays,high mobility scenarios have become increasingly common.The widespread adoption of High-speed Rail(HSR)in China exemplifies this trend,while more promising use cases,such as vehicle-to-everything,continue to emerge.However,the Internet access provided in high mobility environments stllstruggles to achieve seamless connectivity.The next generation of wireless cellular technology 5 G further poses more requirements on the endto-end evolution to fully utilize its ultra-high band-width,while existing network diagnostic tools focus on above-IP layers or below-IP layers only.We then propose HiMoDiag,which enables flexible online analysis of the network performance in a cross-layer manner,i.e.,from the top(application layer)to the bottom(physical layer).We believe HiMoDiag could greatly simplify the process of pinpointing the deficiencies of the Internet access delivery on HSR,lead to more timely optimization and ultimately help to improve the network performance.
基金Supported by the National Natural Science Foundation of China (No.60572093)Specialized Research Fund for the Doctoral Program of Higher Education (20050004016) of China
文摘A novel closed-loop feedback TCP/AQM(Transfer Control Protocol/Active Queue Management) model is proposed in this paper using a discrete-time Markov chain,and a way to calculate the equilibrium distribution of this model is given.In the model,system time is divided into time slots,the bottleneck router queue model and TCP window size model in each slot are analyzed.Finally,by combining adjacent slots,an integrated TCP/AQM analytical model is developed.By this model,the average values of packets dropping ratio and queue length in the router and TCP sending rate can be estimated,hence,this model could be a useful tool for the analysis of AQMs and support the development of new AQM schemes theoretically.The proposed TCP/AQM model is extended to a TCP-UDP(Control User Datagram Protocol)/AQM model,to analyze the TCP/AQM system performance when UDP flows exist.By implementing this model on Matlab,we compare its solutions to NS2 simulated solutions,then the validity of the model to analyze the closed-loop feedback TCP/RED(Random Early Detection) system is verified.
基金supported by the Fundamental Research Fundsfor the Central Universities under Grant No.ZYGX2015J009the Sichuan Province Scientific and Technological Support Project under Grants No.2014GZ0017 and No.2016GZ0093
文摘In data centers, the transmission control protocol(TCP) incast causes catastrophic goodput degradation to applications with a many-to-one traffic pattern. In this paper, we intend to tame incast at the receiver-side application. Towards this goal, we first develop an analytical model that formulates the incast probability as a function of connection variables and network environment settings. We combine the model with the optimization theory and derive some insights into minimizing the incast probability through tuning connection variables related to applications. Then,enlightened by the analytical results, we propose an adaptive application-layer solution to the TCP incast.The solution equally allocates advertised windows to concurrent connections, and dynamically adapts the number of concurrent connections to the varying conditions. Simulation results show that our solution consistently eludes incast and achieves high goodput in various scenarios including the ones with multiple bottleneck links and background TCP traffic.
基金supported by the State Key Program of National Nature Science Foundation of China under Grants No.U0835003,No.60872087
文摘In Wireless Mesh Networks (WMNs),the performance of conventional TCP significantly deteriorates due to the unreliable wireless channel.To enhance TCP performance in WMNs,TCP/LT is proposed in this paper.It introduces fountain codes into packet reorganization in the protocol stack of mesh gateways and mesh clients.Furthermore,it is compatible with conventional TCP.Regarded as a Performance Enhancement Proxies (PEP),a mesh gateway buffers TCP packets into several blocks.It simultaneously processes them by using fountain encoders and then sends them to mesh clients.Apart from the improvement of the throughput of a unitary TCP flow,the entire network utility maximization can also be ensured by adjusting the scale of coding blocks for each TCP flow adaptively.Simulations show that TCP/LT presents high throughput gains over single TCP in lossy links of WMNs while preserving the fairness for multiple TCPs.As losses increase,the transmission delay of TCP/LT experiences a slow linear growth in contrast to the exponential growth of TCP.
基金The National Science Foundation of Chi-na (No.90412010)the Major State Basic Research Devel-opment Program of China(973 Proguam) (No.2003CB317003)
文摘A scheme of setting a limit to the TCP sending window size is proposed to improve the TCP fairness between upload and/or download flows in wired-cum-wireless networks. The goodput and delay of the upload and download TCP flows arc compared to evaluate the TCP fairness for different schemes, which are the different combinations of setting a limit （64 or 4） to the sending window size and using the delayed acknowledgement （ACK） scheme or not. Extensive simulation results and analysis show that （ 1 ） for TCP download flows, setting the limit of sending window size to 4 can improve the fairness; （2） for TCP upload flows, limiting the sending window size and using the delayed ACK strategy are both beneficial to fairness; （3） for TCP download and upload mixture flows, limiting the sending window size to a small value （ e. g. , 4） rather than using the delayed ACK strategy, is the solution to improvement of the fairness ; （4） a large delay interval （200 ms or 300 ms） does not result in improvement in fairness and performance; （ 5 ） a larger TCP packet size （ 1400 B） can improve the TCP upload goodput and decrease the download goodput; in contrast, a smaller TCP packet size （560 B） can increase the download goodput and decrease the upload goodput.
基金This work was financially supported by National Natural Science Foundation of China under grant60273070and60403031,and theNational high-Technology (863) Programunder grant2005AA121560
文摘Network measurement is an important approach to understand network behaviors, which has been widely studied. Both Transfer Control Protocol （TCP） and Interact Control Messages Protocol （ICMP） are applied in network measurement, while investigating the differences between the measured results of these two protocols is an important topic that has been less investigated. In this paper, to compare the differences between TCP and ICMP when they are used in measuring host connectivity, RTT, and packet loss rate, two groups of comparison programs have been designed, and after careful evaluation of the program parameters, a lot of comparison experiments are executed on the Internet. The experimental results show that, there are significant differences between the host connectivity measured using TCP or ICMP; in general, the accuracy of connectivity measured using TCP is 20%- 30% higher than that measured using ICMP. The case of RTT and packet loss rate is complicated, which are related to path loads and destination host loads. While commonly, the RTF and packet loss rate＂ measured using TCP or ICMP are very close. According to the experimental results, some advices are also given on protocol selection for conducting accurate connectivity, RTF and packet loss rate measurements.
