卫星网络中基于蚁群优化的概率路由算法

为了提高空间信息传输的有效性和可靠性,针对传统蚁群优化(ant colony optimization,ACO)容易造成最优路径负载过重而发生拥塞的问题,提出了一种基于蚁群优化的概率路由算法(ant colony optimization based proba-bilistic routing algorithm,ACO-PRA).根据卫星网络拓扑动态周期时变的固有特点,将拓扑周期均匀分为若干个时间片,形成基于不同时间片的卫星网络拓扑连通图;根据网络拓扑连通图,将星间链路带宽和链路容量引入到目标函数中,建立时延最小的优化模型;根据蚁群算法的节点概率函数选择下一跳节点,进而找到一条能同时满足时延带宽和链路容量要求的最佳信号传输路径.仿真结果表明,提出的基于蚁群优化的概率路由算法不仅能够降低平均端到端时延和丢包率,而且能够有效地提高网络吞吐量、平衡网络负载.

Ant colony optimization basedprobabilistic routing algorithm in satellite networks

Aiming at the problem that the traditional ant colony optimization ( ACO) will lead to congestion when the opti-mal path is overloaded, we propose a probabilistic routing algorithm based on ant colony optimization ( ACO-PRA) to im-prove the efficiency and reliability of space information transmission. Firstly, the topology period is divided into several time slices with the consideration of the dynamic and time-varying characteristics for satellite network topology, and then the to-pology connectivity graph is formed based on different time slices. Secondly, according to the topology connectivity graph, bandwidth and link capacity are introduced into the objective function to establish the optimal model of delay minimization. Finally, the next hop node is selected by means of the node probability function of the ant colony algorithm, and then the optimum signal transmission path can be found to meet the requirements of delay bandwidth and link capacity. The simula-tion results show that the proposed ant colony optimization based probabilistic routing algorithm can not only reduce the aver-age end-to-end delay and packet loss rate, but also effectively improve the network throughput and balance the network load.