Field programmable gate array implementation for singular value decomposition in an orthogonal frequency division multiplexing channel estimation
Author:
Affiliation:
Fund Project:
摘要
|
图/表
|
访问统计
|
参考文献
|
相似文献
|
引证文献
|
资源附件
|
文章评论
摘要:
应用基于奇异值分解的OFDM（orthogonal frequency dirision multiplexing）信道估计降低计算量的同时能够保证信道估计的性能，研究OFDM信道向量自相关矩阵奇异值分解的FPGA(field programmable gat array)实现，提出一种复共轭对称矩阵奇异值分解的FPGA简化实现方法，将复矩阵的奇异值分解转换为实矩阵的奇异值分解，并根据实对称矩阵SVD(singular value decomposition)的特点对Systolic阵列结构进行改进，减少了所占用的资源。最后设计了高阶实对称矩阵SVD的FPGA实现方案，仿真结果表明该设计方案是可行的，能够应用于OFDM信道估计系统。
Abstract:
The singular value decomposition (SVD)based orthogonal frequency division multiplexing (OFDM) channel estimation can guarantee channel estimation performance, while reducing computation. A simplified implementation method for SVD of a Hermitian matrix is proposed which transforms a Hermitian matrix SVD to a real symmetry matrix SVD to achieve highspeed operation. The systolic array structure is improved based on the SVD characteristics for symmetry matrix to conserve system resources. The field programmable gate array (FPGA) implementation program of high rank real symmetry matrix singualar value decomposition(SVD)is designed. The simulation results demonstrate that the designed program is feasible for OFDM channel estimation.