Research On Elastic Optical Network Resource Allocation Method Based On Improved Genetic Algorithm

With the development of information technology,exponential growth of various types of business requests has brought tremendous challenges to existing all-optical communication networks.The elastic optical network has become a research hotspot due to its advantages such as fine granularity of spectrum allocation and large bandwidth provision capability.However,while this technology has many advantages,its resource allocation is more complicated due to its flexible resource scheduling.Therefore,this paper studies the Routing and Spectrum Allocation(RSA)problems involved in the allocation of elastic optical network resources.Based on the study of the problem of network resource allocation under pure unicast and single-multicast conditions in optical network services,this paper proposes two algorithms to solve the RSA problem of the elastic optical network under the above two conditions respectively,in order to facilitate the presentation of each The algorithm will be described from two aspects: route calculation and spectrum allocation.1.A Genetic Algorithm Improved Resource Allocation Algorithm(GA-IRAA)is proposed to deal with the allocation of pure unicast service resources.In terms of routing,when the number of requests in the optical network is small with non-blocking service request in the optical network,the GA-IRAA algorithm will focus on finding paths that can achieve higher modulation levels to reduce the use of bandwidth and the load on each link.When there are many requests in the optical network with blocked service requests,the GA-IRAA algorithm will work to reduce the blocking rate.GA-IRAA creates a new random routing path between the source node and the destination node that blocks the service,thus achieving the purpose of reducing the blocking rate through the new random routing.In terms of spectrum allocation,the first hit algorithm is used on the already found routing path.Compared with other algorithms,this algorithm has less computational complexity and lower complexity,and can effectively reduce the blocking rate of late allocation.2.A Genetic Algorithm Hybrid Resource Allocation Algorithm(GA-HRAA)is proposed to deal with the resource allocation of single and multicast mixed services.In terms of routing,a routing path is quickly found for unicast through a fixed alternative routing path used for unicast.The GA-IRAA algorithm is used for multicasting to make the routing path more flat,reducing the spectrum usage width of each link to facilitate distribution and blocking.In terms of spectrum allocation,unicast adopts a distribution method based on spectrum continuity,which ensures rapid allocation while ensuring that the proportion of available bandwidth in the optical network does not drop significantly,leaving space for multicast spectrum allocation.The spectrum distribution method of multicast is the blocking of service classification spectrum,which can ensure the reliability and efficiency of the overall resource allocation.This paper presents a numerical simulation of the effects of the GA-IRAA and GA-HRAA algorithms when dealing with dynamic resource allocation problems.The simulation uses 14 points of the National Science Foundation network topology.The results show that compared with the traditional resource allocation method,the GA-IRAA algorithm has a better performance in reducing the blocking rate under different network traffic,and the rate of decline can reach 3.38% in high network traffic(800Erlang).Compared with the GA-IRAA algorithm,the GA-HRAA algorithm can provide more reliable protection for multicast services.In the case of high network traffic(800Erlang),the blocking rate of multicast services can be reduced by 2.1%.This shows that both algorithms have achieved good results.