Research On The Mechanism Of Fault Location Based On Credibility Model In All Optical Network

Optical networks, driven by the broad band business, tend to be large capacity and high rate. However, these features lead to optical networks extremely easily affected by the faults, which are caused by natural disasters and artificial misoperation. In this case, optical network data loss will be immeasurable once the fault location and network protection can’t be executed accurately and timely.The complexity of optical networks topologies and business deployment, together with the feature of NP-complete make multi-link faults location difficult in transparent optical network. Network manager can’t diagnose the accurate faults number and location without a doubt. The uncertain problems above will be avoided through handling the uncertain relationship between the fault and alarm, which not only increases the accuracy of multi-link faults location, but also guarantees reliable business transfer.Aimed at the uncertain relationship between alarms and fault in all optical networks, the main work related to multi-link faults location in this paper is as follows:First, set up a multi-link fault location model in transparent all optical network with complex topology and business demployment by making the use of credibility reasoning technology, which can turn the uncertainty between the alarms and faults into probability problems efficiently and simplify the multi-link faults location greatly;Second, put forward two heuristic fault location algorithms based on credibility model above, which can measure the credibility of multi-link faults location reasoning process according to alarms quantificationally; Finally, build up a simulation system to simulate the network model and fault propagation model in all optical networks and finish the fault location module to verify the correctness of multi-link faults location model based on credibility and the performance faults location of two algorithms. Simulation results show that the model efficiently solves the uncertainty of fault location, and two algorithms display the good performance on the accuracy of fault location.