DTN中基于贝叶斯的节点相遇概率预测方法  

A Prediction Method of Node Encounter Probability Using Bayesian in DTN

作　　者：马继红[1] 杨文涛[2] 白跃彬[2] 

机构地区：[1]邯郸职业技术学院机电工程系,河北邯郸056001 [2]北京航空航天大学计算机学院,北京100191

出　　处：《计算机测量与控制》2016年第4期185-188,共4页Computer Measurement &Control

基　　金：国家自然科学基金项目(61073076);河北省科技计划支撑项目(13200326D)

摘　　要：DTN网络由于频繁的网络断开、高延迟和异构性等原因,导致网络可用性较低;为了提高DTN网络可用性,一方面要提高数据包送达目的节点的比例,另一方面也要注意控制网络中的副本数量;着重研究在便携设备交换网(PSN)和移动规律性较弱的车载网络(VAN)等网络中对节点相遇概率直接预测的方法;提出了一种利用贝叶斯概率的方法进行相遇概率预测,这种方法基于数据集的历史数据,不依赖于具体的数据集,具有较好的适应性和准确性。Due to the frequent network disconnection, high latency and heterogeneity, DTN network has resulted in low network availability. In order to improve the availability of DTN network, it should increase the proportion of the data packets to the destination node. On the other hand, it also should control the number of copies in the network. The method of direct prediction of node encounter probability in the network, such as PSN, and weak regularity VAN, is studied. It presents a prediction method of Bayesian probability of encounter probability. This method is based on the historical data of data sets, which is not dependent on specific data sets, with better flexibility and accuracy.

关 键 词：DTN网络 贝叶斯网络 节点 相遇概率 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]