Design and implementation of high performance matrix inversion based on reconfigurable processor

IEICE Electronics Express

Jun Lin¹  Feng Han¹  Fan Feng¹  Kun Wang¹  Li Li¹ 

[1] School of Electronic Science and Engineering, Nanjing University

关键词: reconfigurable processor;    matrix inversion;    LU decomposition;    parallel computing;    time-sharing multiplexing;   

DOI  :  10.1587/elex.13.20160579

学科分类：电子、光学、磁材料

来源: Denshi Jouhou Tsuushin Gakkai

【 摘 要 】

References(17)Cited-By(1)In this paper, we propose a high performance matrix inversion implementation on a reconfigurable application specific processor. Our implementation can accelerate variable order matrix inversion ranging from 4 to 144. We adopt LU decomposition to reduce the computation complexity and a pivoting operation to ensure the stability. In order to get higher performance within the limited resources, parallel computing and time-sharing multiplexing are employed. The chip testing results show that our implementation improve the performance of inversion efficiently. The highest parallel speed-up ratio can achieve 3 times, and the execution time of a 144 × 144 matrix inversion is 4.07 ms.

【 授权许可】

Unknown