摘要
兴趣匹配是分布仿真中兴趣管理的关键环节,其速度和精度影响着兴趣管理的性能。基于排序的区域匹配算法是针对HLA/DDM的有效算法,但它只是利用了对范围界点排序而带来的空间关系启发信息,其匹配效率还有值得改进的地方。本文首先分析了匹配算法的实质,然后研究了实际分布仿真系统中实体运行时的局部性,并利用这种局部性对匹配算法进行改进,给出了改进算法,进而对改进算法和原算法进行了比较测试,最后给出了结论和进一步的工作展望。
Interest matching is the key to interest management in distributed simulation. Its speed and precision affects the efficiency of interest management. The sort-based algorithm for region matching is efficient for HLA/DDM. But it only takes the advantage of the heuristic information implied in the location relationships between regions, which can be deduced from the sorted list of range bounds. It can he improved by other means. In this paper, the essence of the sort-based algorithm is analysed first,and then the runtime locality of the simulation entity is studied. And, by using this locality, we present an improved algorithm. After this, comparative tests are taken. Finally, a conclusion and the future trend are given.
出处
《计算机工程与科学》
CSCD
2006年第9期80-82,105,共4页
Computer Engineering & Science
基金
国防预研基金资助项目
