臺灣博碩士論文加值系統

MP3音樂壓縮格式為MPEG-1 Audio Layer3中制定之數位音響編碼標準，由於其優越性已被廣泛應用在網路及無線多媒體之播放與儲存。其中改良型離散餘弦轉換與反轉換(MDCT/IMDCT)是MPEG-1 Layer3中計算最繁複的步驟之一。本論文中我們提出一個遞迴式改良型離散餘弦轉換與反轉換，其核心以無限脈衝響應濾波器(IIR)為基礎的遞迴架構，利用多個濾波器架構來同時完成所有改良型離散餘弦轉換和反轉換的運算。其優點在於它是一個有效率的架構且具有高效能的特性，並具有規則性和模組性之優點。
遞迴式改良型離散餘弦轉換與反轉換非常適合於超大型積體電路之實現，我們將遞迴換演算法在TSMC SPQM 0.35µm製程技術之超大型積體電路設計出來，並利用COMPASS 0.35µm高性能的標準元件製作此架構。晶片佈局面積為 2mm 2mm，使用了7.5k邏輯閘。晶片實際測試結果顯示此晶片之時脈速度可達50MHZ。

The MP3 audio format uses the MPEG-1 audio Layer 3, which is one of the most advanced MPEG (Moving Picture Experts Group) standards for digital audio compression. The MPEG-1 audio Layer 3 has been widely used in internet and wireless communication and storage applications. The modified discrete cosine transform (MDCT) and its inverse transform (IMDCT) are the most complex operations in the MPEG-1 Layer 3 audio coding standard. In this thesis, we propose the new recursive structures for computing MDCT and IMDCT. This recursive structure is based on finite impulse response filters (IIR). We can use several IIR structures for concurrent computing MDCT and IMDCT components. The advantages of the proposed recursive IIR structure are rapid computational efficiency and high throughput rate.
With regularity and modularity, the proposed recursive MDCT and IMDCT figured is suitable for VLSI implementation. In this thesis, we also realize a VLSI chip, which can perform both of the efficient recursion algorithms by using COMPASS 0.35µm high performance cell library with TSMC SPQM 0.35µm process technology. The realized VLSI takes about 7.5k gates in 2mm 2mm silicon area. The measurement result shows that the chip can work at about 50MHz clock rate.

CONTENTS
1.INTRODUCTION…………………………………………………………… 1
1.1Motivation ………………………………………………………………1
1.2Organization of this thesis …………………………………………2
2.OVERVIEW OF THE MPEG AUDIO …………………….…..………4
2.1Motivation ………………………………………………….…..………4
2.2ISO/MPEG audio coding standard …………………………….………4
2.3MPEG-1 Audio …….……………………….…..………………….…6
2.3.1 Encoding …………………………………………………………….8
2.3.2 Decoding …………………………………………………………..10
2.4 MPEG-1 Layers I, II, III coding schemes……………………….10
2.5 MPEG-2………………………………………………………………….15
2.6 MPEG-4 ………………………………………………………………….16
2.7 MPEG audio compression ………………………………………..16
3.RECURSIVE STRUCTURES OF DISCRETE COSINE TRANSFORMAND MODIFIED DISCRETE COSINE TRANSFORM …………………………………………….……………..….19
3.1 Motivation ………….………………………………….……………...….…19
3.2 Discrete cosine transforms ………...…………….…….……….…………...21
3.3 Recursive structures with fixed coefficients ……………………..….22
3.3.1 Fixed recursive structure for DCT-Ⅲ ………………………………23
3.3.2 Fixed recursive structure for DCT-Ⅳ ………………………………28
3.3.3 Fixed recursive structure for DCT-Ⅱ …………………………………32
3.4 Selected and fixed coefficient recursive structures …………………………34
3.4.1 Selected and fixed recursive structure for DCT-Ⅲ ……………..…34
3.4.2 Selected and fixed recursive structure for DCT-Ⅳ ……………..…37
3.5 Recursive structure for MDCT and IMDCT..……………………………….39
3.6 Accuracy and complexity comparisons………….…………………………42
3.7 Conclusion ………………………………………………………………….48
4. RECURSIVE ARCHITECTURE FOR FORWARD AND INVERSE MODIFIED DISCRETE COSINE TRANSFORMS ..49
4.1 Motivation ……………………………………………….…..………….......49
4.2 Related work …………………………….………….……………………....50
4.3 Formulation of recursive MDCT …..…….………….….……………..……53
4.4 Formulation of IMDCT ……………………………………….…………..59
4.5 Implementation and discussion .…………………………………..………67
4.6 Conclusion ………………………………………………………………….67
5. HARDWARE IMPLEMENTATION OF The INVERSE MODIFIED DISCRETE COSINE TRANSFORM BY RECURSIVE STRUCTURE …………….…..………………69
5.1 Motivation …………………………………… ……… …….…..……….69
5.2 The description of IMDCT ………..……………...….…..….……………....70
5.3 Architecture design …………………………………………………….…....75
5.3.1 Pre-processing unit ….…………….………………………….…..……76
5.3.2 Filter banks unit …………………………………...……….……..……76
5.3.3 Post-processing unit ……………………………………………………78
5.4 VLSI implementation of IMDCT ………………………………..……….…79
5.4.1 Design flow ……………………..……………….………………….….79
5.4.2 Simulation result and chip features ………………………...…………..81
5.5 Summary …………….…………………………………………..………….85
6. CONCLUSIONS …………………………………………..…….…..………..86
6.1 Conclusion ………………………………………...…………………………86
6.2 Future work …………………………………………………………………..87
REFERENCES ……………………………… ……..…….………….………..…..88

References
[1] ISO/IEC JTC1/SC29/WG11 MPEG, IS11172-3 “Information Technology—Coding of Moving Picture and Associated Audio for Digital Storage Mediavat up to About 1.5Mbit, Part 3: Audio” 1992. (MPEG-1)
[2] J. P. Princen, A. W. Johnson and A. B. Bradley, “Subband/Transform coding using
filter bank designs based on time domain aliasing cancellation”, Proc. 1987 IEEE
Int. Conf. Acoust. Speech Signal Processing., Dallas, pp. 50.1.1-50.1.4, April 1987.
[3] H. S. Malver, “Lapped transforms for efficient transform/subband coding”, IEEE Trans. Acoust. Speech Signal Processing., vol 38, NO.6 pp. 969-978, June 1990.
[4] L. R. Rao and P. Yip, Discrete Cosine Transform- Algorithms, Advantages Applications, Academic Press, 1990.
[5] N. Ahmeed, T. Natarajan, and K. R. Rao, “ Discrete cosine transform,” IEEE Trans. Comput., vol. C-23, pp. 88-93, Jan. 1974.
[6] R. J. Clarke, “ Relation between the Karhunen-Loeve and cosine transform,” Proc. Inst. Elec. Eng., vol. 128, pt. F, no. 6, pp. 359-360, Nov. 1981.
[7] M. Liou, “Overview of the px64 kbits/s video coding standard,” Commun. ACM, vol. 34, no. 4, pp. 59-63, April, 1991.
[8] G. K. Walllace, “ The JPEG still picture compression standard,” Commun. ACM, vol. 34, no. 4, pp. 30-44, April, 1991.
[9] D. J. LeGall, “ MPEG: a video compression standard for multimeda applications,” Commun. ACM, vol. 34, no. 4, pp. 46-58, April, 1991.
[10] M. Barbero, Succhi, and H. Bailon, “ A flexible architecture for a HDTV codec
based on DCT,” Signal Process. of HDTV Ⅱ, pp. 587-594, L. Chiariglione, ed.
Elsevier, 1990.
[11] S. Wolter, D. Birreck, and R. Laur, “ Classification for 2D-DCTs and a new
architecture with distributed arithmetic,” IEEE Trans. Circuits Syst., vol. 4, pp. 2204-2207, June 1991.
[12] A. Artieri, S. Kritter, F. Jutand, and N. Demassieux “ A one chip VLSI for real time two-dimensional discrete cosine transform,” IEEE ISCAS 1988, pp. 701-704, 1988.
[13] Y. T. Chang and C. L. Wang, “ New systolic array implementation of the 2-D discrete cosine transform and its inverse,” IEEE Trans. Circuits Syst. Video Technol., vol. 5. no. 2, April, 1995.
[14] M. T. Sun, T. C. Chen, and A. M. Gottlieb, “ VLSI implementation of a 16x16 discrete cosine,” IEEE Trans. Circuies Syst., vol. 36, no. 4, pp. 610-617, April, 1989.
[15] J. F. Yang and C. P. Fan, "Recursive discrete cosine transforms with selectable fixed-coefficient filters," IEEE Trans. Circuits Syst.-Ⅱ: Analog and Digital Signal Processing, vol. 46, No. 2, pp. 211-216, Feb. 1999.
[16] Y. F. Jang et al., “ A 0.8u 100Mhz 2D DCT core processor,” IEEE Trans. Consumer Electronics, vol. 40, no. 3, Aug. 1994.
[17] J. C. Carlach et al., “ TCDA: A 27 MHz 8x8 discrete cosine transform chip,” ICASSP 89, pp. 2429-2432.
[18] E. Feig and S. Winograd, "Fast algorithms for the discrete cosine transform," IEEE Trans. Signal Process., vol. 40, pp. 2174-2193, Sept. 1992.
[19] G. Goertzel, "An algorithm for the calculation of finite trigonometric series," Amer. Math. Monthly, vol. 65, pp. 34-35, 1958.
[20] L. P. Chau and W. C. Siu, "Recursive algorithm for the discrete cosine transform with general length," Electron. Lett., vol 30, no. 3, Feb. 1994, pp. 197-198.
[21] Z. Wang, G. A. Jullien, and W. C. Miller, "Recursive algorithms for the forward and inverse discrete cosine transform with arbitrary length," IEEE Signal Processing Lett., vol. 1, no. 7, pp. 101-102, July 1994.
[22] Y. H. Chan, L. P. Chau, and W. C. Siu, "Efficient implementation of discrete cosine transform using recursive filter structure," IEEE Trans. Circuits Syst. Video Technol., vol. 4, no. 6, pp. 550-552, Dec. 1994.
[23] M. F. Aburdene, J. Zheng, and R. J. Kozick, "Computation of discrete cosine transform using Clenshaw’s recurrence formula," IEEE Signal Processing Lett., vol. 2, no. 8, pp. 155-156, Aug. 1995.
[24] A. Y. Wu and K. J. R. Liu, "A low-power and low-complexity DCT/IDCT VLSI architecture based on backward Chebyshev recursion," in Proc. 1994 IEEE Int.Symp. Circuits and Syst., May 1994, vol.4, pp.155-158.
[25] A. Y. Wu and K. J. R. Liu, "Algorithm-based low-power transform coding architectures: the multirate approach," IEEE Trans. VLSI Syst., vol. 6, no. 4, pp. 707-718, Dec. 1998.
[26] K. J. R. Liu, C. T. Chiu, R. K. Kolagotla, and J. F. Jala, " Optimal unified architectures for the real-time computation of time-recursive discrete sinusoidal transforms," IEEE Trans. Circuits Syst. Video Technol., vol. 4, no. 2, pp. 168-180, Apr. 1994.
[27] D. Y. Chan, J. F. Yang, and S. Y. Chen, “Regular implementation algorithms of time domain aliasing cancellation,” IEE Proc. Vis. Image Signal Process., vol. 143, no. 6, pp. 387-392, Dec. 1996.
[28] J. L. Wang, C. B. Wu, B. D. Liu, and J. F. Yang, ”Implementation of the discrete cosine transform and its inverse by recursive structures,” in Proc. SiPS’99, vol. 1, Oct. 1999, pp. 120-130.
[29] H. C. Chiang and J. C. Liu, “Regressive implementation for the forward and inverse MDCT in MPEG audio Coding,” IEEE Signal Processing Lett., vol. 3, no. 4, pp. 116-118, Apr. 1996.
[30] J. Wang and Z. Dong , “A fast algorithm for modified discrete cosine transform,” in Proc. Int. Conf. Commun. Technol., May 1996, vol. 1, pp.445-448.
[31] W. Press, S. A. Teukolsky, W. T. Vetterling, and B. P. Flannery, Numerical Recipes in C: The Art of Scientific Computing. Cambridge, UK: Cambridge University Press, 1992, 2nd ed.