论文部分内容阅读
典型的容迟网络(DTN)场景常表现出大延迟、易中断、高误码率等特点,其中高效节能的路由算法更是一个亟待解决的问题.现有方法主要是通过发送多个副本来提高数据传输的可达性概率,但网络开销很大.为了同时兼顾成功投递率、网络延迟和网络开销,文中提出了基于Markov位置预测模型的DTN路由算法(喷射转发算法).该算法根据节点经过路径的历史信息,用2阶Markov预测机制预测目的节点可能的位置,针对该位置进行多路径的贪婪转发,使包有方向地扩散,减少网络中包的副本数.采用多副本转发的混合发送模式,在保证成功投递率的基础上,有效地减少包副本数,弥补了使用单一模式时不能兼顾网络开销和成功投递率的不足.仿真结果显示,在小节点密度、节点移动速度较快的网络环境下,与spray and wait算法相比,喷射转发算法能有效地提高成功传输率,减小网络开销.
The typical DTN scene often shows great delay, easy to interrupt, high bit error rate, etc. Among them, the efficient and energy-saving routing algorithm is an urgent problem to be solved.The existing methods are mainly to send multiple copies Improve the reachability probability of data transmission, but the network overhead is very large.In order to consider the successful delivery rate, network delay and network overhead at the same time, a DTN routing algorithm based on Markov location prediction model (jet forwarding algorithm) is proposed in this paper, According to the historical information of the path, a second-order Markov prediction mechanism is used to predict the possible location of the destination node, and multi-path greedy forwarding is performed on the location to make the packet spread in the direction and reduce the number of packets in the network. Which can effectively reduce the number of packet copies on the basis of ensuring the successful delivery rate and make up for the lack of network overhead and successful delivery rate when the single mode is used.The simulation results show that in the node density, the nodes move faster Compared with the spray and wait algorithm, the injection-forwarding algorithm can effectively improve the successful transmission rate and reduce the network overhead.