摘要
设计了最短路径时间复杂度取决于边数e和点数n的动态优化算法。采用了独特的动态PV集合链,改进了当前求得的最短路径向量D的存储结构,用PV集合链对向量D进行动态管理,使其时间开销为e+(n-1)×(n-2)/2+3n。当n>4时,SPD OA算法的性能明显优于Dijkstra算法,呈现出良好的动态优化特性。最后对动态优化算法与Dijkstra算法用理论公式得出的数据进行了时间性能比较。
This paper designs the Shortest Path Dynamic Optimum Algorithm(SPDOA) whose time complexity is decided by.the union of the number of edge and the number of vertex,adopts a unique link of PV aggregate,mends the storage structure of currently computing the Shortest Path D vector,makes use of link of PV aggregate to dynamically manage D vector,to attain a spending of time which equals to e+(n-1)×(n-2)/2+3n.While n〉4,the performance of SPDOA is distinctly better than the ones of Dijkstra,the dynamic characteristic is excel.At last,this paper draws comparisons between SPDOA and Dijkstra with the formulas of theory.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第3期82-85,共4页
Computer Engineering and Applications
关键词
最短路径
动态优化算法
PV集舍
单源点
Dijkstra,the shortest path,dynamic optimum algorithm, PV aggregate