针对移动自组网(Mobile Ad Hoc Network,MANET)中路由的广播风暴问题,提出了基于速度感知的可靠的概率路由发现(Speed-aware-based reliable probabilistic route discovery,SRPR)方案。SRPR方案利用节点的速度矢量信息,将节点划分为可...针对移动自组网(Mobile Ad Hoc Network,MANET)中路由的广播风暴问题,提出了基于速度感知的可靠的概率路由发现(Speed-aware-based reliable probabilistic route discovery,SRPR)方案。SRPR方案利用节点的速度矢量信息,将节点划分为可靠节点和非可靠节点。再利用贪婪转发策略,将可靠节点集中选择离目标节点最近的节点赋予高的转发概率,进而提高路由的可靠性,降低路由跳数。将SPRP方案应用于典型的按需式距离矢量路由协议(Ad Hoc On-demand Distance Vector,AODV)进行仿真。仿真结果表明,提出的SPRP方案能够有效缓解广播风暴问题,降低控制包(Routing requests,RREQ)的重传次数,并减少碰撞率。展开更多
To meet the bandwidth requirement for the multicasting data flow in ad hoc networks, a distributed on- demand bandwidth-constrained multicast routing （BCMR） protocol for wireless ad hoc networks is proposed. With th...To meet the bandwidth requirement for the multicasting data flow in ad hoc networks, a distributed on- demand bandwidth-constrained multicast routing （BCMR） protocol for wireless ad hoc networks is proposed. With this protocol, the resource reservation table of each node will record the bandwidth requirements of data flows, which access itself, its neighbor nodes and hidden nodes, and every node calculates the remaining available bandwidth by deducting the bandwidth reserved in the resource reservation table from the total available bandwidth of the node. Moreover, the BCMR searches in a distributed manner for the paths with the shortest delay conditioned by the bandwidth constraint. Simulation results demonstrate the good performance of BCMR in terms of packet delivery reliability and the delay. BCMR can meet the requirements of real time communication and can be used in the multicast applications with low mobility in wireless ad hoc networks.展开更多
Mobile cloud computing （MCC） has become a promising technique to deal with computation- or data-intensive tasks. It overcomes the limited processing power, poor storage capacity, and short battery life of mobile dev...Mobile cloud computing （MCC） has become a promising technique to deal with computation- or data-intensive tasks. It overcomes the limited processing power, poor storage capacity, and short battery life of mobile devices. Providing continuous and on-demand services, MCC argues that the service must be available for users at anytime and anywhere. However, at present, the service availability of MCC is usually measured by some certain metrics of a real-world system, and the results do not have broad representation since different systems have different load levels, different deployments, and many other random factors. Meanwhile, for large-scale and complex types of services in MCC systems, simulation-based methods （such as Monte- Carlo simulation） may be costly and the traditional state-based methods always suffer from the problem of state-space explosion. In this paper, to overcome these shortcomings, fluid-flow approximation, a breakthrough to avoid state-space explosion, is adopted to analyze the service availability of MCC. Four critical metrics, including response time of service, minimum sensing time of devices, minimum number of nodes chosen, and action throughput, are def＇med to estimate the availability by solving a group of ordinary differential equations even before the MCC system is fully deployed. Experimental results show that our method costs less time in analyzing the service availability of MCC than the Markov- or simulation-based methods.展开更多
文摘针对移动自组网(Mobile Ad Hoc Network,MANET)中路由的广播风暴问题,提出了基于速度感知的可靠的概率路由发现(Speed-aware-based reliable probabilistic route discovery,SRPR)方案。SRPR方案利用节点的速度矢量信息,将节点划分为可靠节点和非可靠节点。再利用贪婪转发策略,将可靠节点集中选择离目标节点最近的节点赋予高的转发概率,进而提高路由的可靠性,降低路由跳数。将SPRP方案应用于典型的按需式距离矢量路由协议(Ad Hoc On-demand Distance Vector,AODV)进行仿真。仿真结果表明,提出的SPRP方案能够有效缓解广播风暴问题,降低控制包(Routing requests,RREQ)的重传次数,并减少碰撞率。
基金The Natural Science Foundation of Zhejiang Province(No.Y1090232)
文摘To meet the bandwidth requirement for the multicasting data flow in ad hoc networks, a distributed on- demand bandwidth-constrained multicast routing （BCMR） protocol for wireless ad hoc networks is proposed. With this protocol, the resource reservation table of each node will record the bandwidth requirements of data flows, which access itself, its neighbor nodes and hidden nodes, and every node calculates the remaining available bandwidth by deducting the bandwidth reserved in the resource reservation table from the total available bandwidth of the node. Moreover, the BCMR searches in a distributed manner for the paths with the shortest delay conditioned by the bandwidth constraint. Simulation results demonstrate the good performance of BCMR in terms of packet delivery reliability and the delay. BCMR can meet the requirements of real time communication and can be used in the multicast applications with low mobility in wireless ad hoc networks.
基金Project supported by the National Natural Science Foundation of China （Nos. 61402127 and 61370212） and the Natural Science Foundation of Heilongjiang Province, China （No. F2015029）
文摘Mobile cloud computing （MCC） has become a promising technique to deal with computation- or data-intensive tasks. It overcomes the limited processing power, poor storage capacity, and short battery life of mobile devices. Providing continuous and on-demand services, MCC argues that the service must be available for users at anytime and anywhere. However, at present, the service availability of MCC is usually measured by some certain metrics of a real-world system, and the results do not have broad representation since different systems have different load levels, different deployments, and many other random factors. Meanwhile, for large-scale and complex types of services in MCC systems, simulation-based methods （such as Monte- Carlo simulation） may be costly and the traditional state-based methods always suffer from the problem of state-space explosion. In this paper, to overcome these shortcomings, fluid-flow approximation, a breakthrough to avoid state-space explosion, is adopted to analyze the service availability of MCC. Four critical metrics, including response time of service, minimum sensing time of devices, minimum number of nodes chosen, and action throughput, are def＇med to estimate the availability by solving a group of ordinary differential equations even before the MCC system is fully deployed. Experimental results show that our method costs less time in analyzing the service availability of MCC than the Markov- or simulation-based methods.
