摘要
针对现有可靠性分析方法不能适应卫星网络中信息传播时延长,业务需求多样,导致计算准确性低的问题。考虑卫星工作的多状态特性,研究了多状态下卫星网络可靠性的问题,提出了一种基于QoS的卫星网络k端可靠性分析算法。首先,进行节点融合,对非融合元素采取置零取非操作,进行网络拓扑邻接矩阵变换,得到连通k个节点的路径。相较于传统方法有效地减少了矩阵变换次数,避免了冗余链路的产生。然后,为提升可靠度计算的准确性,根据不同业务的QoS约束条件和链路当前状态,确定网络中满足条件的可用路径,最后,由链路当前状态的可靠度计算出不同业务下的k端路径可靠度。研究结果表明:相较于其他算法,本算法不仅提升了卫星网络在不同业务下k端路径可靠度计算的准确性,而且计算效率提升了33.3%。
Existing reliability analysis methods cannot solve the problem of low calculation accuracy due to prolonged information transmission time in the satellite network and diverse service requirements.Considering the multi-state characteristics of satellite operation,reliability of satellite network in multi-states is studied,and a k-terminal reliability analysis algorithm is proposed based on Quality of Service(QoS).Firstly,the nodes are fused,and the non-fused elements are set to zero;the network topology adjacency matrix is transformed to get the path connecting K nodes.Compared with the traditional method,the matrix transformation is effectively reduced and the redundant links are avoided.Then,to improve the accuracy of reliability calculation,according to the QoS constraints of different services and the current state of the link,the available paths satisfying the conditions in the network are determined.Finally,the k-terminal path reliability for different services is calculated from the reliability of the current state of the link.The results show that compared with other algorithms,the algorithm proposed can improve not only the accuracy of k-terminal path reliability calculation for different services,but also the calculation efficiency by 33.3%.
作者
蔡睿妍
刘艳红
魏德宾
CAI Ruiyan;LIU Yanhong;WEI Debin(Key Laboratory of Communication and Network,Dalian University,Dalian 116622,China;College of Information Engineering,Dalian University,Dalian 116622,China;College of Automation,Nanjing University of Science and Technology,Nanjing 210094,China)
出处
《航空学报》
EI
CAS
CSCD
北大核心
2022年第1期532-541,共10页
Acta Aeronautica et Astronautica Sinica
基金
国家自然科学基金(61901080)。
关键词
卫星网络
多状态
QOS约束
k端可靠性
矩阵变换
satellite network
multistate
quality of service constraint
k-terminal reliability
matrix transformation