在本論文，我們提出一個嵌入式JPEG影像編碼之軟硬體協同傳輸最佳化設計，由底層的硬體設計，至驅動程式到應用程式的軟體設計以及作業系統的開發移植。在硬體設計上，我們將JPEG編碼硬體化，並在匯流排與JPEG編碼硬體之間加入硬體介面電路的設計，作為匯流排與JPEG編碼硬體的溝通操作介面。在軟體設計上，主要有應用程式設計、JPEG驅動程式設計、作業系統移植，其中應用程式包含JPEG解碼、JPEG硬體呼叫和JPEG介面API的設計，JPEG驅動程式主要用來控制JPEG編碼硬體的操作，並且我們將Embedded Linux移植在嵌入式系統中。藉由系統效能分析，我們發現在大量的影像處理系統中，在軟硬體間的資料傳輸佔用了系統時間約30%，為了改善軟硬體資料傳輸的效能，我們利用DMA的存取和記憶體映射的技巧，增加匯流排使用率及減少CPU的操作負擔，並且透過映射技巧，有效減少記憶體搬移的額外負載。我們利用各種不同的影像輸入來分析軟體與硬體各個部份的執行時間，以及分析各種資料傳輸最佳化方法的效能。經由實驗結果分析，我們提出的最佳化嵌入式軟硬體系統在大量影像處理系統上與一般嵌入式系統比較，能得到更佳的效能。

In this paper, we propose an embedded JPEG image compression transfer optimization with software and hardware co-design. The system develop include hardware design and software design with device driver and application .We have porting operation system to our embedded platform. On hardware design, we develop JPEG encoder hardware , then design a communication interface circuit between the JPEG encoder hardware and system bus, communication interface circuit process the bus data receives and JPEG encoder enable. On software design, it can be divided into three segments. The application contains the JPEG decoder, JPEG encoder hardware calling, and a JPEG interface design. The JPEG device driver is used to control the JPEG encoder hardware operation. We will port the Embedded Linux on our embedded system. By system performance estimate , JPEG image compression system the data transfer between the hardware and software takes up about 30% of the total system time, The data translation performance improve between hardware and software . We use the DMA access and memory mapping techniques that increasing bus usage and reducing the CPU overhead . The memory mapping techniques effectively reduce the extra memory moving operation. We use different images to analyze performance that software and hardware process time, and analyzing the our data translation method’s performance. By experimental results analyzing, we propose the embedded hardware and software optimization system greater than general embedded system for large image processing systems.

[1]J. H. Kelm and S. S. Lumetta, "HybridOS: runtime support for reconfigurable accelerators," presented at the Proceedings of the 16th international ACM/SIGDA symposium on Field programmable gate arrays, Monterey, California, USA, 2008.
[2]C. Nitsch, et al., "Embedded system architecture design based on real-time emulation," in Rapid System Prototyping, 2000. RSP 2000. Proceedings. 11th International Workshop on, 2000, pp. 228-233.
[3]J. H. Kelm, et al., "CIGAR: Application Partitioning for a CPU/Coprocessor Architecture," in Parallel Architecture and Compilation Techniques, 2007. PACT 2007. 16th International Conference on, 2007, pp. 317-326.
[4]J. Noseworthy and M. Leeser, "Efficient use of communications between an FPGA's embedded processor and its reconfigurable logic," presented at the Proceedings of the 2006 ACM/SIGDA 14th international symposium on Field programmable gate arrays, Monterey, California, USA, 2006.
[5]G. Madl, et al., "Formal performance evaluation of AMBA-based system-on-chip designs," presented at the Proceedings of the 6th ACM & IEEE International conference on Embedded software, Seoul, Korea, 2006.
[6]X. Zhihui, et al., "A platform-based SoC hardware/software co-design environment," in Computer Supported Cooperative Work in Design, 2004. Proceedings. The 8th International Conference on, 2004, pp. 443-448 Vol.2.
[7]O. A. Nava and A. D. Perez, "Acceleration of Fractal Image Compression Using the Hardware-Software Co-design Methodology," in Reconfigurable Computing and FPGAs, 2009. ReConFig '09. International Conference on, 2009, pp. 167-171.
[8]R. Saleh, et al., "System-on-Chip: Reuse and Integration," Proceedings of the IEEE, vol. 94, pp. 1050-1069, 2006.
[9]ARM Staff , "RealView Platform Baseboard for ARM926EJ-S User Guide , " 2003-2007.
[10]ARM Staff , "AMBA™ Specification , " 2000-2003 Rev 2.0 , " 1999
[11]ARM Staff , "RealView LT-XC4VLX100+ Logic Tile User Guide , " 2006-2007.
[12]JPEG Specification, "Information technology – Digital compression and coding of continuous-tone still images - Requirements and guidelines," ITU-T Recommendation T.81, ISO/IEC 10918-1, 1993
[13]ARM Staff , "PrimeCell DMA Controller Technical Reference Manual , " 2000-2003.
[14]L. Chung-Jr, et al., "Design and implementation of JPEG encoder IP core," in Design Automation Conference, 2001. Proceedings of the ASP-DAC 2001. Asia and South Pacific, 2001, pp. 29-30.
[15]D.P. Bovet and M. Cesati , "UNDERSTANDING THE LINUX KERNEL,3e " O’REILLY , 2006
[16]J. Corbet , A. Rubini and G.K. Hartman. "LINUX DEVICE DRIVERS,3e " O’REILLY , 2005
[17]Wen-Hsiung Chen, C. Harrison Smith, and S. C. Fralick, " A Fast Computational Algorithm for the Discrete Cosine Transform," IEEE Trans. Commun., vol. COM-25, pp. 1004-1009, Sept 1977.
[18]M. Elbadri, et al., "Hardware support of JPEG," in Electrical and Computer Engineering, 2005. Canadian Conference on, 2005, pp. 812-815.
[19]戴顯權 , ｢資料壓縮｣ , 出版商：紳藍. 出版日期： 2002 .
[20]ARM , http://www.arm.com/
[21]Xilinx , http://www.xilinx.com/
[22]Linux Source , http://www.kernel.org/