产生高斯随机序列的新方法

提出了在计算机上产生高斯随机序列的新方法，改进与扩展了通用随机数发生软件的质量与功能．首先，对产生正态分布随机数的Marsaglia-Bray传统方法进行了改进，并用以产生具有良好高斯分布的随机序列；然后，在均方误差最小的准则下，应用双随机交换最小化方法对高斯分布的随机序列进行白色化处理．应用该文提出的方法产生的两个标准高斯随机序列分别具有良好的宽带和窄带白色性能，从而扩展了通用随机数发生软件的功能．

A Novel Method for Generating Gaussian Stochastic Sequences

A novel method for generating standard Gaussian stochastic sequences on computers was presented, which improves and extends the quality and functions of general- purpose software for generating random numbers. Firstly, the conventional Masarglia-Bray algorithm for generating standard normally distributed stochastic sequences is improved, which is then used to generate a random stochastic sequence with a good property of normal distribution. Secondly, a double stochastic interchange minimization algorithm is used to whiten the Gaussian sequence under the criterion of minimum mean-square error. With this method, two standard Gaussian stochastic sequences are generated successfully that have excellent wide-band and narrow-band white characteristics, respectively. Thus, it really extends the functions of general-purpose software for generating random numbers.