In opportunistic networks, most existing buffer management policies including scheduling and passive dropping policies are mainly for routing protocols. In this paper, we proposed a Utility-based Buffer Management str...In opportunistic networks, most existing buffer management policies including scheduling and passive dropping policies are mainly for routing protocols. In this paper, we proposed a Utility-based Buffer Management strategy(UBM) for data dissemination in opportunistic networks. In UBM, we first design a method of computing the utility values of caching messages according to the interest of nodes and the delivery probability of messages, and then propose an overall buffer management policy based on the utility. UBM driven by receivers completely implements not only caching policies, passive and proactive dropping policies, but also scheduling policies of senders. Simulation results show that, compared with some classical dropping strategies, UBM can obtain higher delivery ratio and lower delay latency by using smaller network cost.展开更多
In real complex systems, the limited storage capacity of physical devices often results in the loss of data. We study the effect of buffer size on packet loss threshold in scale-free networks. A new order parameter is...In real complex systems, the limited storage capacity of physical devices often results in the loss of data. We study the effect of buffer size on packet loss threshold in scale-free networks. A new order parameter is proposed to characterize the packet loss threshold. Our results show that the packet loss threshold can be optimized with a relative small buffer size. Meanwhile, a large buffer size will increase the travel time. Furthermore, we propose a Buffered-Shortest-Path-First(BSPF) queuing strategy. Compared to the traditional First-In-First-Out(FIFO) strategy, BSPF can not only increase the packet loss threshold but can also significantly decrease the travel length and travel time in both identical and heterogeneous node capacity cases. Our study will help to improve the traffic performance in finite buffer networks.展开更多
Hypertext transfer protocol(HTTP) adaptive streaming(HAS) plays a key role in mobile video transmission. Considering the multi-segment and multi-rate features of HAS, this paper proposes a buffer-driven resource manag...Hypertext transfer protocol(HTTP) adaptive streaming(HAS) plays a key role in mobile video transmission. Considering the multi-segment and multi-rate features of HAS, this paper proposes a buffer-driven resource management(BDRM) method to enhance HAS quality of experience(QoE) in mobile network. Different from the traditional methods only focusing on base station side without considering the buffer, the proposed method takes both station and client sides into account and end user's buffer plays as the drive of whole schedule process. The proposed HAS QoE influencing factors are composed of initial delay, rebuffering and quality level. The BDRM method decomposes the HAS QoE maximization problem into client and base station sides separately to solve it in multicell and multi-user video playing scene in mobile network. In client side, the decision is made based on buffer probe and rate request algorithm by each user separately. It guarantees the less rebuffering events and decides which HAS segment rate to fetch. While, in the base station side, the schedule of wireless resource is made to maximize the quality level of all access clients and decides the final rate pulled from HAS server. The drive of buffer and twice rate request schemes make BDRMtake full advantage of HAS's multi-segment and multi-rate features. As to the simulation results, compared with proportional fair(PF), Max C/I and traditional HAS schedule(THS) methods, the proposed BDRM method decreases rebuffering percent to 1.96% from 11.1% with PF and from 7.01% with THS and increases the mean MOS of all users to 3.94 from 3.42 with PF method and from 2.15 with Max C/I method. It also guarantees a high fairness with 0.98 from the view of objective and subjective assessment metrics.展开更多
We investigate the similarities and differences among three queue rules,the first-in-first-out(FIFO)rule,last-in-firstout(LIFO)rule and random-in-random-out(RIRO)rule,on dynamical networks with limited buffer size.In ...We investigate the similarities and differences among three queue rules,the first-in-first-out(FIFO)rule,last-in-firstout(LIFO)rule and random-in-random-out(RIRO)rule,on dynamical networks with limited buffer size.In our network model,nodes move at each time step.Packets are transmitted by an adaptive routing strategy,combining Euclidean distance and node load by a tunable parameter.Because of this routing strategy,at the initial stage of increasing buffer size,the network density will increase,and the packet loss rate will decrease.Packet loss and traffic congestion occur by these three rules,but nodes keep unblocked and lose no packet in a larger buffer size range on the RIRO rule networks.If packets are lost and traffic congestion occurs,different dynamic characteristics are shown by these three queue rules.Moreover,a phenomenon similar to Braess’paradox is also found by the LIFO rule and the RIRO rule.展开更多
Evidences indicate that,due to the limited caching capacity or inaccurate estimation on users’preferences,the requested files may not be fully cached in the network edge.The transmissions of the un-cached files will ...Evidences indicate that,due to the limited caching capacity or inaccurate estimation on users’preferences,the requested files may not be fully cached in the network edge.The transmissions of the un-cached files will also lead to duplicated transmissions on backhaul channels.Buffer-aided relay has been proposed to improve the transmission performance of the un-cached files.Because of the limited buffer capacity and the information asymmetric environment,how to allocate the limited buffer capacity and how to incentivize users in participating buffer-aided relay have become critical issues.In this work,an incentive scheme based on the contract theory is proposed.Specifically,the backlog violation probability,i.e.,the buffer overflow probability,is provided based on the martingale theory.Next,based on the backlog violation probability,the utility functions of the relay node and users are constructed.With the purpose to maximize the utility of the relay node,the optimal contract problem is formulated.Then,the feasibility of the contract is also demonstrated,and the optimal solution can be obtained by the interior point method.Finally,numerical results are presented to demonstrate effectiveness of the proposed contract theory scheme.展开更多
In the complex multicore chip system,network on-chip(NoC)is viewed as a kind of system interconnection that can substitute the traditional interconnect networks,which will improve the system performance and communicat...In the complex multicore chip system,network on-chip(NoC)is viewed as a kind of system interconnection that can substitute the traditional interconnect networks,which will improve the system performance and communication efficiency.With regard to the complex and large scale NoC,simple and efficient routing nodes are the critical factors to achieve low-cost and low-congestion communication performance.This paper proposes an unbuffered switch architecture and makes detailed analysis of the mechanism of buffer in the switch architecture.According to the simulation results,the S-mesh using the unbuffered switch architecture is better in terms of the optimal performance in message latency than some typical NoC architectures,such as 2D-mesh,Fat-tree,Butterfly,Octagon and so on.The synthesis results of design compiler indicate that the unbuffered switch has obvious advantages of achieving cost and operating speed for the chips.展开更多
A Shared Multi-buffer Banyan Network is presented in this letter. Its control algorithm and switching fabric are simple, and it fits the high speed ATM network well. The simulation results show that the throughput of ...A Shared Multi-buffer Banyan Network is presented in this letter. Its control algorithm and switching fabric are simple, and it fits the high speed ATM network well. The simulation results show that the throughput of the proposed model is high.展开更多
Jitter is one of the most important issues for multimedia real time services in future mobile ad hoc networks（MANET）. A thorough theoretical analysis of the destination buffer for smoothing the jitter of the real ti...Jitter is one of the most important issues for multimedia real time services in future mobile ad hoc networks（MANET）. A thorough theoretical analysis of the destination buffer for smoothing the jitter of the real time service in MANET is given. The theoretical results are applied in moderate populated ad hoc networks in our simulation, the simulation results show that by predicting and adjusting destination buffer in our way, Jitter will be alleviated in large part and this will contribute much to the quality of service （QOS） in MANET.展开更多
基于OpenFlow的SDN(Software Defined Networking)技术在数据中心中得到广泛研究和应用,如何缓解集中的控制平面成为网络性能的瓶颈是其中的研究热点.OpenFlow规范提出,当数据平面有缓存能力时,未命中的报文仅需发送少量摘要信息至控制...基于OpenFlow的SDN(Software Defined Networking)技术在数据中心中得到广泛研究和应用,如何缓解集中的控制平面成为网络性能的瓶颈是其中的研究热点.OpenFlow规范提出,当数据平面有缓存能力时,未命中的报文仅需发送少量摘要信息至控制器触发规则下发,从而减少控制平面与数据平面的通信负载.然而,现有的缓存模型采用报文粒度的缓存方式,使得同一条流的多个未命中报文会被送至控制器造成额外的通信负载,而且交换机处理报文的顺序会导致流内报文乱序,从而降低通信的性能.针对上述问题,该文提出了一种支持流内报文保序的OpenFlow交换机流缓存管理模型.通过基于流粒度的未命中报文缓存方式,进一步减少控制平面与数据平面的通信开销.通过设计流动作预处理机制,实现同一条流内报文传输保序.该文分别基于软件交换机OFSoftSwitch与硬件网络实验平台NetMagic对该流缓存管理模型进行了原型系统验证.展开更多
We present a staggered buffer connection method that provides flexibility for buffer insertion while designing global signal networks using the tile-based FPGA design methodology. An exhaustive algorithm is used to an...We present a staggered buffer connection method that provides flexibility for buffer insertion while designing global signal networks using the tile-based FPGA design methodology. An exhaustive algorithm is used to analyze the trade-off between area and speed of the global signal networks for this staggered buffer insertion scheme, and the criterion for determining the design parameters is presented. The comparative analytic result shows that the methods in this paper are proven to be more efficient for FPGAs with a large array size.展开更多
基金supported by the National Natural Science Fund of China under Grant No. 61472097the Education Ministry Doctoral Research Foundation of China (20132304110017)the International Exchange Program of Harbin Engineering University for Innovation-oriented Talents Cultivation
文摘In opportunistic networks, most existing buffer management policies including scheduling and passive dropping policies are mainly for routing protocols. In this paper, we proposed a Utility-based Buffer Management strategy(UBM) for data dissemination in opportunistic networks. In UBM, we first design a method of computing the utility values of caching messages according to the interest of nodes and the delivery probability of messages, and then propose an overall buffer management policy based on the utility. UBM driven by receivers completely implements not only caching policies, passive and proactive dropping policies, but also scheduling policies of senders. Simulation results show that, compared with some classical dropping strategies, UBM can obtain higher delivery ratio and lower delay latency by using smaller network cost.
基金Project supported by the National Key Research and Development Program of China(Grant No.2016YFC0802508)the National Natural Science Foundation of China(Grant Nos.11672289 and 61503355)the support from the Chinese Scholarship Council
文摘In real complex systems, the limited storage capacity of physical devices often results in the loss of data. We study the effect of buffer size on packet loss threshold in scale-free networks. A new order parameter is proposed to characterize the packet loss threshold. Our results show that the packet loss threshold can be optimized with a relative small buffer size. Meanwhile, a large buffer size will increase the travel time. Furthermore, we propose a Buffered-Shortest-Path-First(BSPF) queuing strategy. Compared to the traditional First-In-First-Out(FIFO) strategy, BSPF can not only increase the packet loss threshold but can also significantly decrease the travel length and travel time in both identical and heterogeneous node capacity cases. Our study will help to improve the traffic performance in finite buffer networks.
基金supported by the 863 project (Grant No. 2014AA01A701) Beijing Natural Science Foundation (Grant No. 4152047)
文摘Hypertext transfer protocol(HTTP) adaptive streaming(HAS) plays a key role in mobile video transmission. Considering the multi-segment and multi-rate features of HAS, this paper proposes a buffer-driven resource management(BDRM) method to enhance HAS quality of experience(QoE) in mobile network. Different from the traditional methods only focusing on base station side without considering the buffer, the proposed method takes both station and client sides into account and end user's buffer plays as the drive of whole schedule process. The proposed HAS QoE influencing factors are composed of initial delay, rebuffering and quality level. The BDRM method decomposes the HAS QoE maximization problem into client and base station sides separately to solve it in multicell and multi-user video playing scene in mobile network. In client side, the decision is made based on buffer probe and rate request algorithm by each user separately. It guarantees the less rebuffering events and decides which HAS segment rate to fetch. While, in the base station side, the schedule of wireless resource is made to maximize the quality level of all access clients and decides the final rate pulled from HAS server. The drive of buffer and twice rate request schemes make BDRMtake full advantage of HAS's multi-segment and multi-rate features. As to the simulation results, compared with proportional fair(PF), Max C/I and traditional HAS schedule(THS) methods, the proposed BDRM method decreases rebuffering percent to 1.96% from 11.1% with PF and from 7.01% with THS and increases the mean MOS of all users to 3.94 from 3.42 with PF method and from 2.15 with Max C/I method. It also guarantees a high fairness with 0.98 from the view of objective and subjective assessment metrics.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.71801066 and 71431003)the Fundamental Research Funds for the Central Universities of China(Grant Nos.PA2019GDQT0020 and JZ2017HGTB0186)
文摘We investigate the similarities and differences among three queue rules,the first-in-first-out(FIFO)rule,last-in-firstout(LIFO)rule and random-in-random-out(RIRO)rule,on dynamical networks with limited buffer size.In our network model,nodes move at each time step.Packets are transmitted by an adaptive routing strategy,combining Euclidean distance and node load by a tunable parameter.Because of this routing strategy,at the initial stage of increasing buffer size,the network density will increase,and the packet loss rate will decrease.Packet loss and traffic congestion occur by these three rules,but nodes keep unblocked and lose no packet in a larger buffer size range on the RIRO rule networks.If packets are lost and traffic congestion occurs,different dynamic characteristics are shown by these three queue rules.Moreover,a phenomenon similar to Braess’paradox is also found by the LIFO rule and the RIRO rule.
基金the National Natural Science Foundation of China(No.61702258)the Key Projects of Natural Science Research in Colleges and Universities of Jiangsu Province(No.19KJA410001)the Foundation of Jiangsu Advanced Numerical Control Technology Key Laboratory(No.SYKJ201901).
文摘Evidences indicate that,due to the limited caching capacity or inaccurate estimation on users’preferences,the requested files may not be fully cached in the network edge.The transmissions of the un-cached files will also lead to duplicated transmissions on backhaul channels.Buffer-aided relay has been proposed to improve the transmission performance of the un-cached files.Because of the limited buffer capacity and the information asymmetric environment,how to allocate the limited buffer capacity and how to incentivize users in participating buffer-aided relay have become critical issues.In this work,an incentive scheme based on the contract theory is proposed.Specifically,the backlog violation probability,i.e.,the buffer overflow probability,is provided based on the martingale theory.Next,based on the backlog violation probability,the utility functions of the relay node and users are constructed.With the purpose to maximize the utility of the relay node,the optimal contract problem is formulated.Then,the feasibility of the contract is also demonstrated,and the optimal solution can be obtained by the interior point method.Finally,numerical results are presented to demonstrate effectiveness of the proposed contract theory scheme.
基金Supported by the National High Technology Research and Development Program of China(No.2009AA01Z105)the Ministry of EducationIntel Special Foundation for Information Technology(No.MOE-INTEL-08-05)the Postdoctoral Science Foundation of China(No.20080440942,200902432)
文摘In the complex multicore chip system,network on-chip(NoC)is viewed as a kind of system interconnection that can substitute the traditional interconnect networks,which will improve the system performance and communication efficiency.With regard to the complex and large scale NoC,simple and efficient routing nodes are the critical factors to achieve low-cost and low-congestion communication performance.This paper proposes an unbuffered switch architecture and makes detailed analysis of the mechanism of buffer in the switch architecture.According to the simulation results,the S-mesh using the unbuffered switch architecture is better in terms of the optimal performance in message latency than some typical NoC architectures,such as 2D-mesh,Fat-tree,Butterfly,Octagon and so on.The synthesis results of design compiler indicate that the unbuffered switch has obvious advantages of achieving cost and operating speed for the chips.
文摘A Shared Multi-buffer Banyan Network is presented in this letter. Its control algorithm and switching fabric are simple, and it fits the high speed ATM network well. The simulation results show that the throughput of the proposed model is high.
文摘Jitter is one of the most important issues for multimedia real time services in future mobile ad hoc networks（MANET）. A thorough theoretical analysis of the destination buffer for smoothing the jitter of the real time service in MANET is given. The theoretical results are applied in moderate populated ad hoc networks in our simulation, the simulation results show that by predicting and adjusting destination buffer in our way, Jitter will be alleviated in large part and this will contribute much to the quality of service （QOS） in MANET.
文摘基于OpenFlow的SDN(Software Defined Networking)技术在数据中心中得到广泛研究和应用,如何缓解集中的控制平面成为网络性能的瓶颈是其中的研究热点.OpenFlow规范提出,当数据平面有缓存能力时,未命中的报文仅需发送少量摘要信息至控制器触发规则下发,从而减少控制平面与数据平面的通信负载.然而,现有的缓存模型采用报文粒度的缓存方式,使得同一条流的多个未命中报文会被送至控制器造成额外的通信负载,而且交换机处理报文的顺序会导致流内报文乱序,从而降低通信的性能.针对上述问题,该文提出了一种支持流内报文保序的OpenFlow交换机流缓存管理模型.通过基于流粒度的未命中报文缓存方式,进一步减少控制平面与数据平面的通信开销.通过设计流动作预处理机制,实现同一条流内报文传输保序.该文分别基于软件交换机OFSoftSwitch与硬件网络实验平台NetMagic对该流缓存管理模型进行了原型系统验证.
文摘We present a staggered buffer connection method that provides flexibility for buffer insertion while designing global signal networks using the tile-based FPGA design methodology. An exhaustive algorithm is used to analyze the trade-off between area and speed of the global signal networks for this staggered buffer insertion scheme, and the criterion for determining the design parameters is presented. The comparative analytic result shows that the methods in this paper are proven to be more efficient for FPGAs with a large array size.
