[1]闫保中,刘泥石.一种有效的MDCT/IMDCT快速算法[J].应用科技,2011,38(03):12-14.[doi:10.3969/j.issn.1009-671X.2011.03.003]
 YAN Baozhong,LIU Nishi.An efficient algorithm for computation of MDCT/IMDCT[J].Applied science and technology,2011,38(03):12-14.[doi:10.3969/j.issn.1009-671X.2011.03.003]
点击复制

一种有效的MDCT/IMDCT快速算法(/HTML)
分享到:

《应用科技》[ISSN:1009-671X/CN:23-1191/U]

卷:
第38卷
期数:
2011年03期
页码:
12-14
栏目:
自动化技术
出版日期:
2011-03-05

文章信息/Info

Title:
An efficient algorithm for computation of MDCT/IMDCT
文章编号:
1009- 671X(2011)03- 0012- 04
作者:
闫保中刘泥石
(哈尔滨工程大学 自动化学院,黑龙江 哈尔滨,150001)
Author(s):
YAN Baozhong LIU Nishi
(College of Automation, Harbin Engineering University, Harbin 150001, China)
关键词:
音频编码语音编码改进型离散余弦变换快速算法
Keywords:
audio coding speech coding modified discrete cosine transform fast algorithm
分类号:
TP391.41
DOI:
10.3969/j.issn.1009-671X.2011.03.003
文献标志码:
A
摘要:
提出了一种输入序列长度为N=5×2m的改进型的离散余弦变换(MDCT)的有效算法,可以有效减少数据量,提高计算机储存和运算效率.首先将序列长度为N的MDCT转化为N/2的离散余弦变换IV型(DCT-IV) ,然后将后者转化为长度为N/2的离散余弦变换II型(DCT-II),最后再通过加法和乘法运算实现快速计算过程.同时,分析该算法的算术复杂度.结果表明,较之传统方法,提出的算法能使常用窗型下的MDCT算术复杂度降低20%以上,实现了音频和语音编码领域的运算效率的提高.
Abstract:
A novel and efficient algorithm for the computation of modified discrete cosine transform (MDCT) with block size N=5×2m is proposed, which can effectively reduce data volume and therefore improve the efficiency of storage and computing. Firstly, the MDCT with block size N is converted into type-IV discrete cosine transform (DCT-IV) which is further transformed into DCT-II with block size N/2. Secondly, a modified window is shown, which is piece-wise symmetric. Finally, addition and multiplication operations are conducted for rapidly computing the DCT-II. Analysis of the proposed algorithm in terms of computational complexity is provided. The proposed algorithm achieves at least 20% reduction of arithmetic operations compared to the traditional method, which is meaningful for audio and speech coding.

参考文献/References:

[1]PRINCEN J P, BRADLEY A B. Analysis/synthesis filter bank design based on time domain aliasing cancellation [J]. IEEE Trans Acoustics Speech Signal Process, 1986, 34(5): 1153-1161.
[2]BRITANAK V, RAO K R. An efficient implementation of the forward and inverse MDCT in MPEG audio coding [J]. IEEE Signal Process Lett, 2001,8(2):48-51.
[3]LEE S W. Improved algorithm for efficient computation of the forward and backward MDCT in MPEG audio coder [J]. IEEE Trans Circuits Syst II, 2001,48(10): 990-994.
[4]李琳, 郭立,白雪飞,等. 基于N/8点FFT核的MDCT/IMDCT快速实现方案[J]. 中国科学院研究生院学报, 2008, 25(6): 810-815.
[5]周茜, 伍家松, 董志芳,等. 一种新的三维MDCT的快速算法[J]. 东南大学学报:自然科学版, 2010,40(1): 75-79.
[6]BRITANAK V, RAO K R. A new fast algorithm for the unified forward and inverse MDCT/MDST computation[J]. Signal Processing, 2002,82(3): 433- 459.
[7]HEIDEMAN M T. Computation of an odd-length DCT from a real-valued DFT of the same length[J]. IEEE Transactions on Signal Processing, 1992,40(1):54-61.
[8]CHENG M H, HSU Y H. Fast IMDCT and MDCT algorithms——a matrix approach[J]. IEEE Transactions on Signal Processing, 2003,51(1):221-229.
[9]KOK C W. Fast algorithm for computing discrete cosine transform[J]. IEEE Transactions on

备注/Memo

备注/Memo:
作者简介:闫保中(1963-), 男,研究员,硕士生导师,主要研究方向:组合导航,E-mail:heuybz@126.com.
更新日期/Last Update: 2011-04-12