CAJ | 학술논문

综合利用节点物理状态信息(位置和速度矢量信息)和社会关系信息(历史相遇信息),提出一种多信息融合的概率路由算法( PRMF)。该算法依据节点间距离的远近程度动态调整两类信息在预测相遇关系中的权重,以选择合适的信息来计算传输概率,并依据传输概率的大小选择转发节点。另外,还引入有效的分组副本控制和分组散播策略,并结合预测信息改进了节点缓存管理方法。仿真结果表明：与现有的几种路由算法相比,PRMF保持了较低的平均开销,同时具有最高的投递率和最小的平均延迟。
종합이용절점물리상태신식(위치화속도시량신식)화사회관계신식(역사상우신식),제출일충다신식융합적개솔로유산법( PRMF)。해산법의거절점간거리적원근정도동태조정량류신식재예측상우관계중적권중,이선택합괄적신식래계산전수개솔,병의거전수개솔적대소선택전발절점。령외,환인입유효적분조부본공제화분조산파책략,병결합예측신식개진료절점완존관리방법。방진결과표명：여현유적궤충로유산법상비,PRMF보지료교저적평균개소,동시구유최고적투체솔화최소적평균연지。
With an integrated utilization of nodes'physical status information and social relationship infor-mation,i. e. ,position and velocity vector information and historical encounter information,a probabilistic routing algorithm based on multi-information fusion,called PRMF,is proposed for delay tolerant networks ( DTN) in this paper. According to the relative distance from a node to the destination node,PRMF dynami-cally adjusts the weighting factor to select the appropriate information for calculation of the node's delivery probability which gives guidance to forwarder selection. In addition, an effective strategy for packet copy control and packet spreading is also introduced in PRMF to improve the queue management method by combining with the former predictive information. The simulation results show that PRMF can have not only a relatively small average overhead but also a higher packet delivery rate and a lower delivery delay than other routing algorithms.