以优化的链路状态路由(optimized link state routing,OLSR)协议为原型,为无线Mesh网设计了一种基于节点区分和跨层设计的优化链路状态路由协议NDCL-OLSR(node-differentia-tion and cross-layer based optimized link state routing)....
详细信息
以优化的链路状态路由(optimized link state routing,OLSR)协议为原型,为无线Mesh网设计了一种基于节点区分和跨层设计的优化链路状态路由协议NDCL-OLSR(node-differentia-tion and cross-layer based optimized link state routing).该协议采用了一种新的路由判据NDCLM(node-differentiation and cross-layer metric),在路由计算过程中通过跨层操作机制综合考虑了节点当前负载和链路投递率2个因素对链路质量的影响,并且对节点的类型进行了区分,使得网络流量尽量绕开业务繁忙的超网关节点(SGW),从而大大优化了路由选择的效果.仿真结果表明,NDCL-OLSR能够从很大程度上提高网络的吞吐量,降低端到端的延时,并且能够达到负载均衡的路由效果.
The model of energy cost in a wireless sensor network （WSN）environment is built, and the energy awareness and the wireless interference mainly due to different path loss models are studied. A special case of a clust...
详细信息
The model of energy cost in a wireless sensor network （WSN）environment is built, and the energy awareness and the wireless interference mainly due to different path loss models are studied. A special case of a clustering scheme, a twodimensional grid clustering mechanism, is adopted. Clusterheads are rotated evenly among all sensor nodes in an efficient and decentralized manner, based on the residual energy in the battery and the random backoff time. In addition to transmitting and receiving packets within the sensors＇ electrical and amplification circuits, extra energy is needed in the retransmission of packets due to packet collisions caused by severe interference. By analysis and mathematical derivation, which are based on planar geometry, it is shown that the total energy consumed in the network is directly related to the gridstructure in the proposed grid based clustering mechanism. The transmission range is determined by cluster size, and the path loss exponent is determined by nodal separation. The summation of overall interference is caused by all the sensors that are transmitting concurrently. By analysis and simulation, an optimal grid structure with the corresponding grid size is presented, which balances between maximizing energy conservation and minimizing overall interference in wireless sensor networks.
针对成本约束有向无环图DAG(directed acyclic graph)表示的网格工作流完工时间最小化问题,提出两个基于优先级规则的迭代启发算法.算法利用并行活动特征定义正向分层和逆向分层两个概念,将其分别引入最大收益规则MP(maximum profit),得到正分层最大收益规则MPTL(maximum profit with top level)和逆分层最大收益规则MPBL(maximum profit with bottom level).两规则每次迭代尽量以完工时间的最小增加换取总费用的最大降低,逐步将分层初始解构造为满足成本约束的可行解.模拟结果表明,两规则在获得较少迭代次数和运行时间的同时,能显著改进MP规则的平均性能,且MPBL优于MPTL.
暂无评论