多媒體應用在嵌入式系統中一直扮演著很重要的角色，本篇論文針對低價嵌入式產品提出了一個具有多媒體功能強化的微控器架構(Multimedia Enhanced MCU，簡稱ME-MCU)，藉由在8位元微控器核心中加入了32位元多媒體延伸指令集的方式，有效提升8位元微控器的多媒體訊號處理能力，並且保留了微控器之低成本的特性。除此之外，為了讓CPU設計者能夠快速的發展多媒體延伸架構，我們將原本實作在ME-MCU中的多媒體延伸模組(MME Module)調整成一個可重覆使用的矽智產元件(SIP)，它採用SIMD(Single Instruction Multiple Data)架構有效的提升資料處理的平行度。目前已成功的將MME SIP整合至另外一顆嵌入式微處理器－ARM7架構中，在此個案的實驗中我們得到了平均3.41倍的效能提升，而所增加硬體成本也控制在11%的幅度內，以效能與成本的比值表現來看，本個案算是一個成功的設計。

Multimedia workloads have always held an important role in embedded applications. The thesis presents a Multimedia Enhanced MicroController Unit (ME-MCU) for low-cost embedded applications. The main components of ME-MCU include a PIC16C 8-bit MCU core with a 32-bit multimedia extension, an enhanced memory interface fo signal processing, and an embedded In-Circuit Emulator (ICE). The main goal is not only to improve the media performance, but also to keep the small-area benefit of signle-chip microcontrollers. In addition, base on the architecture of ME-MCU, we proposed a reusable MultiMedia Extension SIP (MME SIP) for providing the CPU designers a simple solution to perform the multimedia extension to an exist ISA. A cost-effective multimedia extension to ARM7 microprocessors has been introduced as a case study of MME SIP. The MME SIP is capable of reducing the execution cycle counts by 69% to 81% for several critical loops, with an average speedup of 3.41, while the gate count overhead is kept under 11%.

Chapter 1. 簡介(Introduction) 1
1.1 背景說明(Background) 1
1.2 研究動機(Motivation) 2
1.3 主要方法(Proposed Method) 2
1.4 本篇論文的貢獻(Contribution) 3
1.5 論文結構(Thesis Organization) 3
Chapter 2. 相關研究(Related Works) 4
2.1 PIC16C 8-bit Microcontroller 4
2.2 ARM7TDMI 32-bit General Purpose Microprocessor 5
2.3 多媒體延伸指令集(Multimedia Instruction Set Extension) 7
Chapter 3. MME延伸指令集(MME Instruction-Set Extension) 10
3.1 單一指令處理多筆資料(Single Instruction Multiple Data；SIMD) 10
3.2 多媒體運算(Multimedia Operations) 11
3.3 Programming Model 13
3.4 指令編碼(Instruction Encoding) 15
Chapter 4. 硬體微架構(Hardware Architecture) 18
4.1 時脈方法與指令週期(Clocking Scheme/Instruction Cycle) 18
4.2 中央處理器之微架構(Micro-architecture of the CPU) 20
4.3 資料記憶體介面(Interface to Data Memory) 25
4.4 週邊模組(Peripheral Modules) 26
4.5 指令集架構之驗證(Instruction-Set Architecture Verification) 28
Chapter 5. 實驗分析與討論(Experiments and Discussions) 32
5.1 硬體成本比較(Comparison of Hardware Cost) 32
5.2 功能/成本之分析(Analysis of Functions and Cost) 33
5.3 效能分析與比較(Performance Analysis and Comparison) 35
5.4 管線組織之探討(Discussion on Pipeline Organization) 46
Chapter 6. 應用方法(Application Note) 50
6.1 硬體的整合(Hardware Integration) 50
6.2 軟體的撰寫(Software Design) 51
6.3 應用實例(Application Example) 52
Chapter 7. 多媒體延伸模組之IP化(The Reusable Multimedia Extension Module) 54
7.1 MME IP之功能 54
7.2 MME IP之實作 55
7.3 MME IP之驗證 56
7.4 MME IP之應用 57
Chapter 8. 具多媒體強化之ARM7微處理器(Multimedia Enhanced ARM7 Microprocessor) 59
8.1 延伸指令集的制訂(Definition of Extended Instruction-Set) 59
8.2 硬體架構的調整(Hardware Modification) 60
8.3 效能與成本分析(Analysis of Performance and Cost) 62
Chapter 9. 結論(Conclusion) 64
參考文獻(References) 66
附錄A (Appendix A) – ME-MCU指令編碼表 67
附錄B (Appendix B) – Memory Mapped Registers 71

[1]. J. Eyre and J. Bier, “The Evolution of DSP Processors,” IEEE Signal Processing Magazine, Volume: 17, Issue: 2, March 2000, pp. 43 – 51
[2]. R. Bannatyne and G. Viot, “Introduction to Microcontrollers,” Wescon/98, 1998, pp. 350-366
[3]. Peleg and U. Weiser, “MMX Technology Extension to the Intel Architecture,”IEEE Micro, pp. 42–50, July/Aug. 1996
[4]. M. Tremblay et al., “VIS Speeds New Media Processing,” IEEE Micro, pp. 10–20, July/Aug. 1996
[5]. S. Furber, ARM system-on-chip architecture, 2nd, Addison-Wesley, 2000
[6]. Microchip Technology Inc., “PIC16Cxx Data Sheet,” http://www.microchip.com
[7]. Nintendo Inc., http://www.gameboy.com
[8]. R. B. Lee, “Subword Parallelism with MAX-2,”IEEE Micro, pp. 51–59, July/Aug. 1996
[9]. R. B. Lee, “Multimedia Extensions for General-Purpose Processors,” Proc. IEEE Workshop on Signal Processing Systems, pp. 9–23, Nov. 1997
[10]. D. A. Patterson and J. L. Hennessy, Computer Organization & Design: The Hardware/Software Interface, 2nd, Morgan Kaufmann, 1997
[11]. S. J. Orfanidis, Introduction to Signal Processing, Prentice Hall Inc., 1996
[12]. J. Bergeron, Writing Testbenches Functional Verification of HDL Models, Kluwer Academic Publishers, 2000
[13]. M. G. Arnold, Verilog Digital Computer Design, Algorithms into Hardware, Prentice Hall, 1999
[14]. Virtual IP Group Inc., “C-DSP Data Sheet,” http://www.virtualipgroup.com
[15]. Improv Systems Inc., “Jazz DSP Data Sheet,” http://www.improvsys.com
[16]. H. Yoo, E. Ko, M. Sunwoo, “A Multimedia DSP Chip for Portable Applications,” Proc. the First IEEE Asia Pacific Conference on ASICs, pp. 147-150, 1999
[17]. C. Lee, M. Potkonjak and W. H. Mangione-Smith, “MediaBench: A Tool for Evaluating and Synthesizing Multimedia and Communications Systems”, Proc. International Symposium on Microarchitecture, pp. 330-335, 1997
[18]. I. J. Huang, W. K. Huang and C. F. Kao, “A Parameterized MMX IP Module for RISC Microprocessors”, Proc. the 12th VLSI Design/CAD Symposium, pp. 35, Aug. 2001
[19]. I. J. Huang, W. K. Huang, R. T. Gu and C. F. Kao, “A Cost Effective Multimedia Extension to ARM7 Microprocessors,” Proc. IEEE International Symposium on Circuits and Systems, pp. 304-307, Mar. 2002