一种抑制停滞的蚁群路由算法

An Ant Colony Routing Algorithm with Mitigating Stagnation

蚁群路由算法(ACR)作为生物行为的模拟,在网络中得到了广泛的应用。然而,停滞现象严重影响了它的性能和自适应性。目前的大多数蚁群路由算法对减轻停滞现象的研究主要集中于信息素的处理,较少考虑停滞带来的网络负载失衡。本文提出了一种抑制停滞的蚁群路由算法(MS-ACR),该算法通过计算信息素变化量和改进概率更新函数来预防停滞现象。当停滞出现时,MS-ACR算法使用抑制措施来缓解停滞现象,快速发现新的最优路径。仿真表明,该算法在负载平衡以及丢包等重要的网络性能上具有较为明显的优势。

As the mimic of natural behavior, ant colony routing algorithm （ACR） has been widely used in the practical networks. However, stagnation phenomenon has a serious impact on its performance and adaptivity. At present, the existing approaches to alleviate stagnation mainly concentrate on the pheromone processing and seldom concern the load unbalance. This paper proposes an ant colony routing algorithm with mitigating stagnation （MS-ACR）. With the improvements in computing variety of pheromone and updating probability, MS-ACR prevents stagnation phenomenon and when stagnation already appears, it adopts corresponding mitigating measurements to alleviate stagnation and finds the alternate optimal paths. Computer simulations and analysis indicate that, MS-ACR has advantage of load balance, and some vital routing performances like network packet loss ratio etc. over other algorithms compared.