摘要
在并行FFT(快速傅里叶变换)算法的基础上,进一步研究了FFT算法的并行特点,设计了一种基于矩阵运算的并行FFT算法以及矩阵乘法的DLP(数据级并行)计算实现方法,并利用矩阵乘法的DLP计算实现了并行FFT算法,并在多核仿真平台上实现了新的并行算法。仿真结果表明:较传统算法,新的并行化算法在运行时间上有所提升,接收机的性能有了很大的提高。
This paper presented the characteristics of the parallel FFT( fast Fourier transform) algorithm on the basis of further research on parallel FFT algorithm,designed a parallel FFT algorithm and matrix multiplication algorithm based on DLP matrix( data parallel calculation) method,and calculated the implementation of parallel FFT algorithm based on matrix multiplication DLP,and realized the new parallel algorithm on multi-core platform. The simulation results show that,compared with the traditional algorithm,the new parallel algorithm is improved in running time,and the performance of the receiver is greatly improved.
作者
韩美林
HAN Mei-lin(School of Electronic Information and Electrical Engineering,Shangluo University,Shangluo 726000,Shaanxi Province,China)
出处
《信息技术》
2018年第6期64-67,共4页
Information Technology
基金
2016年商洛市科技局项目(SK2016-52)
关键词
子载波解调算法
并行算法
FFT
阵列处理器
subcarrier demodulation algorithm
parallel algorithm
FFT
array processor