首页 | 本学科首页   官方微博 | 高级检索  
   检索      

作战仿真系统中高性能随机数发生器研究
引用本文:徐享忠,潘丽君.作战仿真系统中高性能随机数发生器研究[J].装甲兵工程学院学报,2009,23(2):67-70.
作者姓名:徐享忠  潘丽君
作者单位:装甲兵工程学院,装备指挥与管理系,北京,100072
摘    要:为满足作战仿真系统对随机数发生器随机性、鲁棒性、周期和效率等的苛刻要求,采用组件技术,开发了素数模乘同余组合发生器。该随机数发生器生成的随机数,其分布类型、参数和独立性经检验满足要求;其周期长达10^18,且不受种子的影响。该随机数发生器采用的算法经过优化后,运算速度提高了近40%。研究成果解决了仿真作战过程随机性的难题,为多样本的并行仿真、仿真结果置信区间的得出奠定了基础,有助于提高仿真系统的可信性。

关 键 词:伪随机数  组合发生器  线性同余法  蒙特卡洛仿真

Study on High Performance Random Number Generators in Operation Simulation Systems
XU Xiang-zhong,PAN Li-jun.Study on High Performance Random Number Generators in Operation Simulation Systems[J].Journal of Armored Force Engineering Institute,2009,23(2):67-70.
Authors:XU Xiang-zhong  PAN Li-jun
Institution:(Department of Equipment Command and Administration, Academy of Armored Force Engineering, Beijing 100072, China)
Abstract:In order to meet the severe requirements of the operation simulation systems in the following aspects, such as randomicity, robustness, period and efficiency etc, a combined Prime Modules Muhiplicarive Linear Congruential Generator (PMMLCG) is developed with component technology. The random number generated, whose distribution type, parameters and independency can meet the requirements through test. Its period is as long as 10^18 , and it doesn't rely on the seeds. The computational efficiency is improved by nearly 40% after the algorithm is optimized. The research results solve the problem of randomicity during operation simulation, and lay a solid foundation for parallel simulation of multiple samples and derivation of confidence interval of simulation results. The results also help improve the credibility of the simulation systems.
Keywords:pseudorandom number  combined random number generators  linear congruential method  Monte Carlo simulation
本文献已被 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号