隨著三維繪圖技術與3C可攜式產品的快速發展，三維圖形繪圖已經廣泛地應用在手持裝置上，例如筆記型電腦、個人數位助理(Personal Digital Assistant, PDA)與智慧型手機等等。一般而言，三維圖形繪圖需要複雜的運算，而大量的運算會快速消耗電池電力，因此滿足高運算量的代價將是顯著縮短電池的供電時間，但手持裝置的電池蓄電量是有限的。再者，莫爾定理(Moore’s law)預測每18個月晶片單位面積中的電晶體數目會增加一倍，但目前電池製造技術的發展遠不及處理器的發展速度。此外，可攜式產品核心處理器的供電電壓隨著製程技術的進步日漸下降，若要兼顧系統效能與電池壽命，將會額外增加電源管理的困難度，所以如何有效管理電源已成為手持裝置產品相當重要的設計關鍵之ㄧ。
對目前三維圖形系統而言，動態調整電壓與頻率(Dynamic Voltage and Frequency Scaling, DVFS)是實現電源管理的主要方法主之一，DVFS須使用一個有效的即時預測方法來預測畫面的工作量，據此適當地調整電壓與頻率以節省多餘的功率消耗。因此，許\多相關文獻皆針對三維圖形系統提出不同的工作量預測方法，目前常見的預測方式有特徵(Signature-based)預測法[1]、歷史訊息(History-based)預測法[3]以及比例積分微分(Proportion-Integral-Derivative, PID)預測法[14]，但大多數方法是將電源管理機制放在軟體層(中央處理器)，如此不但無法快速地預測圖形處理器(Graphic Processing Unit, GPU)的執行時間，同時也造成手持系統中央處理器的運算負擔。
本論文提出一個以比例積分控制器為主控制器、模糊控制器為僕控制器的電源管理預測器，透過模糊控制器的模糊概念調整比例積分控制器中的比例參數，修正傳統PI控制器需要以繁瑣的湊試法來尋找適合的比例、積分參數，強健比例積分控制器的適應性與預測精準度。此外，以Uniform Window-size Predictor 1 (UW1) 預測法為輔，適時地使用UW1預測法能夠使預測跟上實際工作量變化的腳步。實驗數據顯示本預測器比適應性比例積分預測器[18]的預測精準度提升到6.8%，可節省能量約1.4%，而電路面積與也只比原比例控制器硬體多增加1.3%，消耗的功率在GPU中多消耗0.02%。本論文亦將提出的預測器應用於已上市的三維第一人稱遊戲─雷神之錘，其結果印證本預測器確實為一有效的預測方法，它的適應性可以克服真實遊戲的工作量變化，準確地調整電壓頻率，藉此降低功率消耗，達到省電之目的。

As time goes by rapid development of 3D graphics technique and 3C portable product output, 3D graphics have been widely applied to handheld devices, such as notebooks, PDAs, and smart cellular phones. Generally, to process 3D graphics applications in mobile devices, processor needs strong capability of handling large computational-intensive workloads. Complex computation consumes a great quantity of electric power. But the lifetime of handheld device battery is limited. Therefore, the cost, to satisfy this demand, will be shortening the supply time of device battery. Moreover, Moore’ law said that the number of transistors in a chip is double in every eighteen months. But these days the advance in manufacturing batteries still cannot get up with the advance in developing processors. In addition, the improvement of chip size has led to more small, supply voltage of kernel processor in portable device. Considering system efficiency and battery lifetime simultaneously increase the difficulty of designing power management scheme. So, how to manage power effectively has become one of the important key for designing handheld products.
For 3D graphics system, dynamic voltage and frequency scaling (DVFS) is one of good solutions to implement power management policy. DVFS needs an efficient online prediction method to predict the workload of frames and then appropriately adjust voltage and frequency for saving energy consumption. Consequently, a lot of related papers have proposed different prediction policy to predict the executing workload of 3D graphics system. For instance, the existing prediction policies include signature-based[1], history-based[3] and proportion-integral-derivative (PID)[14] methods, but most of designers put power management in software, i.e. processors. This solution not only slows power management to get the information about executing time of graphic processing unit (GPU), but also increases the operating overhead of CPU in handheld system.
In this paper, we propose a power management workload prediction scheme with a framework of using proportion-integral (PI) controller to be a master controller and fuzzy controller to be a slave controller, and then implement it into hardware circuit. Taking advantage of fuzzy conception in fuzzy controller is to adjust the proportional parameter in PI controller, the shortage of traditional PI controller that demands on complicated try-and-error method to look for a good proportional and integral parameters can be avoided so that the adaption and forecasting accuracy can be improved. Besides, Uniform Window-size Predictor 1 (UW1) is also implemented as an assistant manner. Using UW1 predictor appropriately can improve the prediction trend to catch up with the trend of real workload. Experimental results show that our predictor improves prediction accuracy about 3.8% on average and saves about 0.02% more energy compared with PI predictor[18]. Circuit area and power consumption only increases 6.8% percent and 1.4% compared with PI predictor. Besides, we also apply our predictor to the 3D first person game, Quake II, in the market. The result shows that our predictor is indeed an effective prediction policy. The adaption can put up with the intense workload variation of real game and adjust voltage and frequency precisely to decrease power consumption and meet the purpose of energy saving.

第一章 緒論 1
1.1 研究動機 1
1.2 論文大綱與架構 2
第二章 理論背景 4
2.1 電源管理策略簡介 4
2.2 低功率設計技術簡介 8
2.2.1. 時脈閘控 9
2.2.2. 運算元隔離 9
2.2.3. 電源閘控 10
第三章 三維電腦繪圖系統晶片 11
3.1 系統架構與繪圖流程 11
3.2 電源管理方法 14
3.3 預測策略 15
3.3.1. 特徵(Signature-based)預測法 15
3.3.2. 歷史訊息(History-based)預測法 17
3.3.3. 比例-積分-微分(PID)預測法 18
3.3.4. 模糊(Fuzzy)預測法 21
第四章 混合型預測器設計 26
4.1 比例積分預測器 26
4.2 適應型模糊比例積分預測器 27
4.3 C模組 34
4.4 預測流程 35
4.5 硬體電路 37
4.6 Benchmarks 39
4.6.1. API Test Bench 39
4.6.2. FPGA Test Bench 42
4.6.3. Quake II 42
4.7 環境設定與評估公式 43
4.7.1. 電壓頻率對應表 43
4.7.2. 評估公式 45
4.8 實驗結果 46
4.8.1. 預測器硬體數據 46
4.8.2. C Model模擬結果一 47
4.8.3. C Model模擬結果二 47
4.8.4. C Model模擬結果三 48
4.8.5. PM硬體實現於RTL之模擬結果 48
4.8.6. PM硬體實現於CDK之模擬結果 51
第五章 預測器整合於三維圖形系統晶片 53
5.1 電源管理軟硬體 53
5.2 電源管理之整合 55
5.2.1. 軟體整合─API與PM 55
5.2.2. 硬體整合─PM與CG 55
5.2.3. 硬體整合─PMCG與電壓管理 59
5.2.4. 硬體整合─GMRM+PMCG in RTL 60
5.2.5. 硬體整合─PM RTL與SystemC co-simulation 60
5.2.6. 軟硬體整合─PM-SW與PM-HW 61
第六章 結論與未來工作 63
參考文獻 65

[1] Bill Pontikakis, Hung Tien Bui, Francois-Raymond Boyer, and Yvon Savaria, “A Low-Complexity High-Speed Clock Generator for Dynamic Frequency Scaling of FPGA and Standard-Cell Based Designs,” IEEE International Symposium on Circuits and Systems, pp.633-636, 2007
[2] Bren C. Mochocki, Kanishka Lahiri, Srihari Cadambi, and X. Sharon Hu, “Signature-Based Workload Estimation for Mobile 3D Graphics,” Design Automation Conference, pp.592-597, 2006.
[3] Bren Mochocki, Kanishka Lahiri and Srihari Cadambi, “Power Analysis of Mobile 3D Graphics,” IEEE Design Automation Conference, pp.592-597, 2006
[4] Byeong-Gyu Nam, Jeabin Lee, Kwanho Kim, Seungjin Lee and Hoi-Jun Yoo, “Cost-Effective Low-Power Graphics Processing Unit for Handheld Devices,” IEEE on Communications Magazine, Vol. 46, pp.152-159, 2008
[5] E.H. Mamdani, “Application of fuzzy logic to approximate reasonling using linquistic synthesis,” IEEE Trans. Computers, vol. c-26, 1977
[6] G. A. Paleologo, L. Benini, A. Bogliolo and G. De Micheli, “Policy Optimization for Dynamic Power Management,” IEEE Design Automation Conference, pp.182-187, 1998
[7] Hyunjin Kim, Hyejeong Hong, Hong-Sik Kim, Jin-Ho Ahn and Sungho Kang, “Total Energy Minimization of Real-Time Tasks in an On-Chip Multiprocessor Using Dynamic Voltage Scaling Efficiency Metric,” IEEE Transactions on Computer-aided Design of Integrated Circuits and Systems, Vol. 27, No. 11, pp.2088-2091, 2008.
[8] Jin-Lin Lin, Kun-Yi Wu, and Shiann-Rong Kuang, “Low Power Mapping and Pipelined Scheduling Using Tabu Search,” National Computer Symposium, 2007
[9] P.M. Larsen, “Industrial applications of fuzzy logic control,” Int. J.Man, Mach, Studies, vol. 12, no. 1, pp. 3-10, 1980
[10] Richard Golding and Peter Bosch, “Idleness is not sloth,” USENIX 1995 Technical Conference, pp.17-17, 1995
[11] Sung-Yong Bang, Kwanhu Bang, Sungroh Yoon and Eui-Young Chung, “Run-Time Adaptive Workload Estimation for Dynamic Voltage Scaling,” IEEE Transactions on Computer-aided Design of Integrated Circuits and Systems, Vol. 28, No. 9, pp.1334-1346, 2009
[12] Yung-Hsiang Lu and Giovanni De Micheli, “Comparing system level power management policies,” IEEE Design & Tests, Vol. 18, pp.10-19, 2001
[13] Yan Gu, Samarjit Chakraborty and Wei Tsang Ooi, “Games are Up for DVFS,” IEEE Design Automation Conference, pp.598-603, 2006
[14] Yan Gu and Samarjit Chakarborty, “Control Theory-based DVS for Interactive 3D Games,” IEEE/ACM Design Automation Conference (DAC), pp.740-745, 2008
[15] 許華山, “三維圖形系統晶片電源管理策略之設計與實作,” 國立中山大學資訊工程學系碩士論文, 2008
[16] 周景揚, “CICeNEWS,” National Chip Implementation Center, Vol. 78, 2007
[17] 魏慶隆, “CICeNEWS,” National Chip Implementation Center, Vol. 86, 2007
[18] 鄭浩逸, “應用在三維圖形系統晶片電源管理之適應性比例積分控制器”, 國立中山大學資訊工程學系碩士論文, 2009
[19] 陳家偉, “適應性類神經補償之PID控制系統,” 中華技術學院機電光工程研究所碩士論文, 2005
[20] “自動控制系統中PID整定,” www.symcukf.com