并行射线跟踪算法及其在城市电波预测的应用  

Parallel ray-tracing algorithm and its application for propagation prediction in urban microcellular environments

作　　者：刘海涛[1] 黎滨洪[1] 谢勇[1] 戚冬生[1] 

机构地区：[1]上海交通大学电子工程系,上海200030

出　　处：《电波科学学报》2004年第5期581-585,共5页Chinese Journal of Radio Science

摘　　要：射线跟踪算法的计算量较大 ,耗时较长。针对这一问题提出并研究了对等模式和主从模式两类并行射线跟踪算法 ,结合MPI并行运算函数库 ,在局域网计算机簇中 ,实现了城市复杂微蜂窝环境的电波预测。结果表明 ,并行算法在精度相同的情况下 ,大大缩减了计算机运行时间。而且 。Ray-tracing algorithm needs a great deal of computation time. To improve its efficiency, two kinds of parallel ray-tracing algorithm: Peer to Peer Model (PPM) and Master/Slaves Model (MSM) are presented. Propagation prediction simulation in urban microcellular environments is made in LAN workstation cluster by using MPI parallel functions library. The comparison among numeric results indicates that the parallel algorithm considerably reduces the computation time without losing precision, and that the MSM can achieve better parallel speedup gain and load balancing than PPM.

关 键 词：并行射线跟踪 电波预测 MPI 并行加速增益 负载均衡 

分 类 号：TN011[电子电信—物理电子学]