针对稀疏分布有限波长转换能力的光突发交换(SWCC-OBS)网络容易出现的高丢包率和负载不均衡问题,建立了稀疏分布OBS全网丢包率的数学模型。为了克服模型求解过程中多变量带来的非线性问题,对影响全网突发丢包率的参变量进行分解预估和反馈更新,求得全网对应业务流模型下的优化路由路径,获得了全网突发数据的近似最低丢包率。仿真实验表明,在SWCC—OBS网络中,本文提出的路由选择模型及分解协调的路由选择算法,不仅能有效地避免突发数据包间的相互冲突、降低了全网丢包率,而且能很好地适应突发数据业务的动态变化,保证了网络的负载均衡和较低的丢包率。
Optical burst switching (OBS) has been proposed as a promising switching technology for the next generation of optical transport networks. Particular attention is paid to the commercial OBS networks with sparse wavelength conversion capabilities (SWCC-OBS). An overall burst packet loss probability, which adequately represents the congestion state of entire network, is the primary metric of interest in an SWCC-OBS network. With the overall loss probability as the optimization objective, a novel routing model is proposed based on integer non-linear programming formulation. In Order to overcome the computational complexity, a novel routing algorithm based on decomposition coordination is proposed to solve the problem by decomposing and pre-forecasting the variables. Also it updates the values according to the feekback. Finally, it will obtain an optimal routing path and the global approximate optimal path to minimize the burst loss probability. The simulation results show that the proposed model and algorithm can give a near-optimal routing that avoids the conflict of burst data effectively and decreases the overall burst loss probability. At the same time,it also performs well in balancing payloads throughout the whole network under different kinds of burst data patterns and network topologies.