论文部分内容阅读
为解决容迟网络(DTN)覆盖层资源分配效率受接入带宽概率不确定性影响的问题,将路由资源分配和不确定规划模型相结合,提出针对路径可预测移动环境下的新模型,该环境由物流物联网抽象而来.模型加入不确定理论判定条件,以大于给定值的概率保证资源分配的施行;并使用中断和时延效用函数作为模型的最优化尺度,以遗传算法结合神经网络调整参数得到最佳值,从而提高用户的综合满意度.NS2仿真结果验证其在接入带宽概率不确定环境下效率更高;吞吐率相较于未采用不确定规划的情形可提高8.30%～15.4%,业务量较大的时候传输整体性能更为突出.在搭配有效的路径预测算法后,该模型可以扩展到一般环境.
To solve the problem that the resource allocation efficiency of DTN overlay is affected by the uncertainty of access bandwidth, a new model for routing predictable mobile environment is proposed by combining routing resource allocation and uncertain programming model. The environment is abstracted from the logistics internet of things.The model joins the judgment conditions of uncertainty theory to ensure the allocation of resources with a probability greater than a given value.Using the interrupt and delay utility function as the optimization measure of the model, Network adjustment parameters to get the best value, so as to improve the user’s comprehensive satisfaction.NS2 simulation results verify that it is more efficient under the condition of uncertain access bandwidth probability; the throughput rate can be increased by 8.30% compared with the case without using the uncertain programming, ~ 15.4%, the overall performance of transmission is more prominent when the traffic is large, and the model can be extended to the general environment after the effective path prediction algorithm is provided.