摘要
针对一些典型分布的噪声信号在雷达标定和信号处理中的实际需求,基于元胞自动机和Box-Muller算法提出一种在现场可编程门阵列上能够实时产生多种概率分布特性的噪声信号生成方法。首先将对数和三角函数序列存入只读存储器;然后使用两个32阶的元胞自动机,产生两路服从均匀分布的随机序列,作为地址读取存储器的数据进行相应乘法运算,实现Box-Muller算法的快速运算,生成两路正交的高斯随机序列;最后利用高斯随机序列进行相应数学运算得到其他概率分布特性的随机序列。实验使用Modelsim进行了仿真,并结合MATLAB做了验证,仿真结果表明,随机序列统计直方图与标准分布曲线高度吻合,误差小于0.1%且正交性良好。基于Xil-inx的XC7Z100FFG900现场可编程门阵列和ADI的数模转换器AD9779进行实现,得到的实验结果与仿真一致,且可以通过上位机对生成信号速率和类型进行切换,速度最高可达4 Gb/s,满足实际需求。
To solve the practical requirements of some typically distributed noise signals in radar calibration and signal processing,a noise signal generation method that can generate a variety of randomly distributed characteristics in real time on a Field Programmable Gate Array(FPGA)based on cellular automata and Box-Muller algorithm was proposed.First,logarithmic,and trigonometric sequences are stored in Read-Only Memory(ROM).Then,two 32nd-order cellu-lar automata are used to generate two random sequences that obey uniform distribution,which are used as the data of the address reading memory for corresponding multiplication operations,to realize the fast operation of the Box-Muller algo-rithm and generate two orthogonal Gaussian random sequences.Finally,the Gaussian random sequence is used to per-form the corresponding mathematical operation to obtain the random sequence with other probability distribution charac-teristics.The simulation results show that the statistical histogram of the random sequence is in good agreement with the standard distribution curve,the error is less than 0.1%and the orthogonality is good.The experimental results obtained by the XC7Z100FFG900 field programmable gate array based on Xilinx and AD9779 digital-to-analog converter based on ADI are consistent with the simulation,and the generated signal rate and type can be switched by the host computer,with the speed-up to 4Gb/s,to meet the actual demand.
作者
赵将冬
张福贵
汪治
ZHAO Jiangdong;ZHANG Fugui;WANG Zhi(College of Electronic Engineering,Chengdu University of Information Technology,Chengdu 610225,China;Key Laboratory of Atmospheric Exploration of China Meteorological Administration,Chengdu 610225 China)
出处
《成都信息工程大学学报》
2025年第1期29-35,共7页
Journal of Chengdu University of Information Technology
