Research On Balancing Routing Algorithm For Energy Consumption And QoS In Elastic Optical Networks

Due to the rapid increasing of diversified internet services and multimedia applications,such as high-definition television,3-D video on demand,cloud services and big data,backbone networks are suffering great challenges.The traditional wavelength division multiplexing network is limited due to the fixed and coarse granularity wavelength grid,which causes the wasting of bandwidth resources.Compared to the traditional wavelength division multiplexing network,the elastic optical network based on optical-orthogonal frequency division multiplexing can allocate the number of sub-carriers flexibly according to the request rate.Meanwhile,the elastic optical network can adaptively select different modulation formats according to the transmission distance.Therefore,the elastic optical network is widely considered to be a promising next generation optical transport network.However,with the explosive growth of optical network traffic and network scale,it is bound to lead to a sharp increase in the energy consumption and cost of optical network.The development of energy efficient and flexible optical network is extremely urgent.In addition,the diversified traffic development and user demand of the elastic optical network make the quality of service provided by operators need to meet different parameter constraints,such as blocking rate and reliability,so as to ensure business transmission.Meanwhile,energy consumption and quality of service are mutually restricted performance indicators.Therefore,the energy consumption and quality of service balancing routing algorithm is mainly studied in elastic optical networks.Firstly,for reducing the energy consumption in elastic optical networks and mitigating the increasement of bandwidth blocking probability,a spectrum reservation and load balancing energy efficiency routing strategy is proposed in the third chapter.In this strategy,a load state formula of node is designed to report the link spectrum usage of node-related links on the shortest path of traffic.Then,the node load state can be determined by comparing the spectrum status of node-related links on the shortest path and average spectrum status in the elastic optical network.At the same time,the load state of candidate lightpath is decided by node spectrum state and available frequency slots of candidate lightpath.Available spectrum aware reservation grooming algorithm and load balancing routing algorithm are proposed and performed independently based on load state of candidate path.The available spectrum aware reservation grooming algorithm focuses on energy saving by reserving lightpath dynamically according to available frequency slots of lightpath.While load balancing routing algorithm focuses on bandwidth blocking probability by designing a lightpath cost formula,which is with respect to hops and spectrum continuity.Moreover,a balance index is designed to evaluate the balance between energy efficiency and blocking performance.Compared with the green grooming algorithms of distance-adaptive spectrum resource allocation,simulation results show that the proposed routing strategy can balance the energy consumption and blocking probability significantly.Secondly,in order to address the problems of low spectrum utilization and high energy consumption caused by physical impairment in service diversified elastic optical networks,a service differentiated energy efficiency routing strategy with link impairment-aware spectrum partition is proposed in the fourth chapter to guarantee the reliable transmission of traffic.In order to reduce the nonlinear impairment between different channels,a path weight formula jointly considering the link spectrum state and transmission impairment is designed based on load balancing.Begin with the highest modulation level layered auxiliary graph which is constructed by spectrum efficiency and reach,the link-disjoined maximum weight paths are selected for high quality requests according to path weight formula.The link-disjoined shortest energy efficiency paths are selected for low quality requests in auxiliary graph.During the phase of spectrum allocation,partition spectrum according to the traffic rate,then the first-fit and last-fit spectrum allocation policies are combined to reduce the cross-phase modulation between the requests which have different transmission rates.In order to reduce the effects of the new request provisioning on established traffic,a reconfiguration mechanism is proposed to reconfigure all affected traffic.The simulation results show that the proposed route strategy can improve the bandwidth blocking probability and energy consumption performance.