矢量基二维DCT修剪在DSP上的内存存取减少方法

doi:10.3969/j.issn.0372-2112.2019.03.034

PDF(1209 KB)

电子学报 ›› 2019, Vol. 47 ›› Issue (3) : 757-763. DOI: 10.3969/j.issn.0372-2112.2019.03.034

科研通信

矢量基二维DCT修剪在DSP上的内存存取减少方法

刘项洋, 许勇, 郑孝遥, 陈付龙

作者信息 +

Memory Access Reduction Method for Vector-Radix 2D DCT Pruning on DSP

LIU Xiang-yang, XU Yong, ZHENG Xiao-yao, CHEN Fu-long

Author information +

文章历史 +

摘要

本文针对矢量基二维DCT修剪提出内存存取减少方法.该方法旨在减少计算中因权重因子和信号输入而导致的内存存取.它首先利用权重因子的属性将计算流程图内每相邻两阶段内的蝴蝶运算单元进行融合，然后再以较少的权重因子来计算.本文采用通用DSP处理器来验证该方法对矢量基二维DCT修剪算法的有效性.并且实验结果显示该方法相比于常规方法可以大幅度减少运算所需的时钟周期数、降低对运算中对内存的存取量、以及占用更少的内存.

Abstract

This paper proposes a novel memory reference reduction method for vector-radix 2D DCT pruning.This method aims to reduce the memory reference owing to weighting factors and signal input.The proposed method merges the butterflies at every neighboring two stages in the computation diagram,and then computes them with fewer weighting factors.Hardware platform based on general purpose processor is used to verify the effectiveness of the proposed method for vector-radix 2-D FCT pruning implementation.Experimental results validate the benefits of the proposed method with less clock cycle,less memory reference, and fewer memory space compared with the conventional implementation.

导出引用

刘项洋, 许勇, 郑孝遥, 陈付龙. 矢量基二维DCT修剪在DSP上的内存存取减少方法[J]. 电子学报, 2019, 47(3): 757-763. https://doi.org/10.3969/j.issn.0372-2112.2019.03.034

LIU Xiang-yang, XU Yong, ZHENG Xiao-yao, CHEN Fu-long. Memory Access Reduction Method for Vector-Radix 2D DCT Pruning on DSP[J]. Acta Electronica Sinica, 2019, 47(3): 757-763. https://doi.org/10.3969/j.issn.0372-2112.2019.03.034

中图分类号： TN911.23

参考文献

[1] N AHMED,T NATARAJAN,K R RAO.Discrete CosineTransform[J].IEEE Transactions on Computers,1974,C-23(1):90-93.
[2] K RRao,P Yip.Discrete Cosine Transform:Algorithms,Advantages,Applications[M].New York:Academic,1990.
[3] 龚若皓,杨斌.基于移动多核GPU的并行二维DCT变换实现方法[J].成都信息工程学院学报,2015,30(1):22-26.GONG Ruo-hao,YANG Bin.Parallelization of 2D-DCT based on Mobile Multicore GPU[J].Journal of Chengdu University of Information Technology,2015,30(1):22-26.(in Chinese)
[4] 陈睿,王晶,黄华军,M R Alsharif.基于分块DCT变换的多聚焦图像融合[J].小型微型计算机系统,2016,37(2):321-326.CHEN Rui,WANG Jing,HUANG Hua-jun,M R Alsharif.Multi-focus image fusion based on block DCT transform[J].Journal of Chinese Computer Systems,2016,37(2):321-326.(in Chinese)
[5] 桑爱军,崔新宇,王艇,李晓妮.2M维矢量矩阵DCT整数变换及并行实现[J].东北大学学报(自然科学版),2017,38(11):1543-1547.SANG Ai-jun,CHUI Xin-yu,WANG Ting,LI Xiao-ni.2M-dimensional vector matrix DCT integer transform and parallel implementation[J].Journal of Northeastern University (Natural Science),2017,38(11):1543-1547.(in Chinese)
[6] Giovanni Renda,Maurizio Masera,Maurizio Martina,Guido Masera.Approximate arai DCT architecture for HEVC[A].2017 New Generation of CAS (NGCAS)[C].Genova,Genoa:IEEE,2017.133-135.
[7] 贾昆霖.基于8×8的整数DCT快速计算的H.264/AVC软件实现[J].电子科技,2017,30(6):43-45.JIA Kun-Lin.Realization of H.264/AVC software based on 8×8 integer DCT fast calculation[J].Electronic Science and Technology,2017,30(6):43-45.(in Chinese)
[8] Diego F Coelho,Sushmabhargavi Nimmalapalli,Vassil Dimitrov,Arjuna Madanayake,Renato J Cintra,Arnaud Tisserand.Computation of 2D 8×8 DCT based on the loeffler factorization using algebraic integer encoding[J].IEEE Transactions on Computers,2018,(67)12:1692-1702.
[9] S C Chan,K L Ho.A new two dimensional fast cosine transformalgorithm[J].IEEE transactions on signal processing,1991,39(2):481-485.
[10] E Feig,S Winograd.On the multiplicative complexity of discrete cosine transform[J].IEEE Transactions on Information Theory,1992,38(4):1387-1391.
[11] C A Christopoulos,J Bormans,J Cornelis,A N Skodras.The vector-radix fast cosine transform:Pruning and complexity analysis[J].Signal Processing,1995,43(2):197-205.
[12] Muye Norley Liu.Vector-radix DCT/ICT implementation for MPEG DSP[A].Proceedings of the Third International Conference on Signal Processing[C].Beijing,China:IEEE,1996.641-644.
[13] X Liu,HBao.Efficient implementation of 2-D FCT with reduced memory access for programmable DSPs[J].Journal of Signal Processing Systems,2013,80(2):153-161.
[14] A N Skodras,A G Constantinides.Efficient input-reordering algorithms for fast DCT[J].Electronics Letters,1991,27(21):1973-1975.