Research On Protection Strategy For Space Division Multiplexing Elastic Optical Networks

With the emergence of new Internet applications,the demand for network capacity is growing rapidly.Compared with the traditional wavelength division multiplexing optical network,the elastic optical network using multi-core optical fiber can utilize spectrum resources more finely and flexibly,so that the network capacity can be improved effectively.However,this also brings unprecedented challenges to the survivability of the network.Failure of any fiber in the space division multiplexing optical network will result in the loss of massive business data.Therefore,it is urgent to design a fast and efficient survivability strategy to ensure the network's anti-destructive performance.Among them,the protection strategy has become the preferred method in the optical network survivability strategy with the feature of fast recovery.However,the space division multiplexing elastic optical network has a large capacity characteristic,and the protection strategy will consume a large amount of network resources,at the same time,its unique inter-core crosstalk can also seriously affect network performance.To this end,a machine learning-based differentiated availability anti-disaster protection strategy and a crosstalk-aware service degradation-based protection strategy are proposed for high traffic load condition and low traffic load condition respectively.In the case of low network load,in order to achieve the maximum sharing of protected spectrum resources under the premise of satisfying the service availability requirements,a machine learning-based differentiated availability protection strategy against disasters is purposed.The strategy accurately senses the service availability requirements by the linear discrimination analysis based support vector machine,and prioritizes the shared protection resource allocation according to its availability requirements,and proposes an alternate spectrum core selection scheme to reduce cross-core crosstalk.The simulation results show that compared with the existing strategy,the proposed strategy not only improves the service availability satisfaction rate,but also reduces the spectrum redundancy rate and cross-core crosstalk.Under the high load of the network,due to the shortage of network resources,the above strategy will face the problem of increased blocking rate and crosstalk.In order to ensure the survivability of the network under the premise of satisfying the certain transmission quality requirements of the service,a degraded service protection strategy based on crosstalk-aware is proposed.The strategy uses the decision tree method to classify services and set priorities according to transmission quality requirements.At the same time,the transmission distance and inter-core crosstalk of integrated services are designed to design a prioritized degraded service algorithm.The simulation results show that the proposed strategy can achieve lower blocking rate and spectrum redundancy,while maintaining cross-core crosstalk within an acceptable range.