Research On Node Importance And Protection Strategy Based On High-Speed Railway Optical Transport Network

The high-speed railway optical transport network is the basis for ensuring the safety of railway operation.It can provide accurate and timely information transmission for train status monitoring,operation control,command and dispatching,remote fault diagnosis and etc.In the event of a transmission interruption,it will seriously affect the normal operation of the high-speed railway.In order to avoid the huge damage to the railway operation caused by the network failure,the research of node importance,node failure location and node protection strategy for high-speed railway optical transport network is necessary.The existing node importance algorithms are only considered from a single static or dynamic dimension,and are not applicable to the safe operation of high-speed railway optical transport network.For the node failure location,the focus of previous research has been the rational allocation of monitors,but they are so expensive.It is complex for the existing node P-Cycle protection algorithm to construct the basic P-Cycles.The A prior Efficiency(AE)about node is not used to evaluate and select the node P-Cycles.Therefore,the existing node importance,node failure location and node P-Cycle algorithms are improved in this thesis.The main research results are as follows:(1)A Two Dimensional Evaluation Algorithm(TDEA)is proposed and applied to the weightless and weighted networks,based on the static network dimension and the dynamic network dimension of the node in this thesis.For the static network dimension,the more important the node is,the more impact it has on encircling nodes.According to this,the Node Importance Contribution Matrix(NICM)algorithm is improved.At the same time,the node efficiency and adjacent node contribution rate are selected as the evaluation indexes.For the dynamic network dimension,the traffic through the node is considered as an evaluation index.The simulation results verify the accuracy of the algorithm and the effectiveness of the indexes.And the running time of the TDEA is better than NICM algorithm.(2)For the node failure location,an optical pulse reflector location method is proposed in this thesis.The location method is to place optical pulse reflectors on both sides of each link.After inputting an optical pulse,the cross-correlation function values of the total output signal and the theoretical reflection signal of each pulse reflector are calculated,and the failure location can be accurately located compared with the set threshold.For the node protection,a Tangent P-Cycle Merging Algorithm(TPMA)based on single node failure protection is proposed.The specific steps are as follows:first,the minimum node-encircling P-Cycles are constructed in such a way that all adjacent links of the central node are expanded;then,with the modified A priori Efficicency(m-AE)as a restriction,the m-AE increased tangent P-Cycles are merged;finally,by setting the parameter K,the average m-AE of different hops P-Cycle can be obtained.The typical COST239 network topology is used to simulate,the results show that the TPMA algorithm is better than the existing Node-encircling P-Cycle Mining Algorithm(NCMA)in terms of the minimum P-Cycles construction time,the number of candidate P-Cycles and the average m-AE of the P-Cycles.