一种粒子群优化的用户优先级虚拟网络映射算法

以底层网络资源利用率最大化为目标,对控制转发分离网络建立基于"资源抢占+重映射"的用户优先级虚拟网络映射整数线性规划模型,并提出了一种改进离散粒子群算法来解决虚拟网络映射问题.该算法的粒子进化更具方向性,同时引进不同粒子位置互斥因子,解决粒子群算法易早熟陷入局部最优解的缺陷.最后通过仿真实验从节点资源利用率、链路资源利用率、一般虚拟网络接受率、平均跳数和长期运营收益成本比等方面,将改进离散粒子群算法与贪婪算法和二进制离散粒子群算法对比,验证了改进离散粒子群算法的高性能.

Particle swarm optimization user-priority virtual network embedding algorithm

In the forwarding and control separation network, we model the user-priority virtual network embedding problem as an integer linear programming, which is achieved with resource grabbing and re-mapping aiming at maximizing the resource utilization of the substrate physical network. And we propose a modified discrete particle swarm optimization algorithm (M_DPSO) for short to solve the VN embedding problem. In the M_DPSO, the particle evolves more directionally, and the mutually exclusive factor of different particle positions is introduced to resolve the problem of premature and easily becoming local optimal solution. Finally, the performance parameters, including node resource utilization, link resource utilization, the VN accept rate, the average jump number and the long-term operators benefit cost ratio, are evaluated by emulation experiments. In contrast to the greedy algorithm and binary particle swarm optimization algorithm, the M_DPSO is verified to be of high performance.