基于局部位置信息和消息投递度的受控传染路由算法

Location Position and Message Delivery Ratio Based Controlled Epidemic Routing for DTNs

容延/容断网络(DTN)具有间断连接、时延高和节点缓存受限等特点,因而其路由策略面临巨大的挑战.为了提高容延网络的消息投递率,本文利用位置信息和相遇信息来选择下一跳节点,并基于此提出了受控传染路由算法LPDR.该算法在消息源节点和中继节点上采取不同的策略.在消息源节点上运用局部位置信息,利用节点的局部位置信息来控制消息的扩散范围.在中继节点上,综合利用多种效用信息筛选出最优的节点进行消息复制,从而更好地控制了消息的冗余.仿真实验表明在基于Random Waypoint节点移动模型的网络环境中,与LC-Epidemic、Epidemic、Prophet相比,LPDR的消息投递率最高,平均跳数最少,网络负载比LC-Epidemic和Epidemic分别低约25%和15%.

Delay/Disruption Tolerant Network（ DTN） has characteristics of intermittent connection,long delay and limitation of buffer space. So the routing strategy faces with a great challenge. To improve the message delivery rate,this paper selects the next hop node by using the location information and encounter information. Based on this,we propose a controlled epidemic routing algorithm LPDR.The algorithm adopts different strategies on the message source node and relay node. On the message source node,the local position information is used to control the spread of the message. On the relay node,we use various kinds of utility information to filter out the best node for the message replication,which can better control the message redundancy. Simulation experiments have been conducted and the results demonstrate that LPDR has higher message delivery ratio and less average hop count than other three algorithms and the overhead ratio of LPDR is less than 25% and 15% of LC-Epidemic and Epidemic respectively in a Random Waypoint mobility model based DTN environment.