在低功率的晶片設計中，我們通常都是透過關掉某部分的電路來節省電力的消耗，不同於以往的論文，我們關掉的部份不僅僅只是控制器的部份，也包含的資料處理器的部份，而在延伸式自動狀態機中，狀態機的機率提供了我們運算的執行頻率，且每一個狀態中的運算也提供了我們運算所需要的資源需求，經由資源需求，我們可以來決定共享資源的多少，經由我們的實驗，我們發現約有10%的面積減少和24%的電力減少。
實驗之初，先讀入一個延伸式自動狀態機的模型，將這個模型透過我們的分析及分解，得到一個個小的模型，再分別對每個模型做控制器的合成和資料處理器的合成，最後再交由設計編譯器來得到我們的實驗數據。

不同於以往的論文，只針對控制器或處理器的部份做關掉的動作，我們提出了一個有效的演算法可以根據機率和資源需求來做有效的分解並關掉不用的部份來節省能源，而關掉的部份同時包含的控制器的部份和資料處理器的部份，透過關掉部份的電路，我們可以達到低功率的目的。

在實驗中，我們以我們結果和資源全部共享及全部不共享來做比較，並根據狀態機率和面積來做分析，由實驗結果我們發現我們的想法是正確的，透過我們的方法確定可以達到低功率的目的。

Previous work has shown that power reduction can be achieved through turning off portions of circuit when they are idle. Unlike previous work which focused only on either controller or datapath, we propose a decomposition technique taking both controller and datapath into consideration. The state probability of a FSM provides the execution frequency of operation in that state. Operations performed in states provide the resource requirement which can be used to determine the resource sharing among states. Experimental results show that on the average, 10% area reduction and 24% power reduction can be achieved as compared to designs without decomposition.

[1] M. Sami, D. Sciuto, C. Silvano, V. Zaccaria and R. Zafalon. "Exploit-
ing Data Forwarding to Reduce the Power Budget of VLIW Embedded
Processors", DATE 2001, , 2001.
[2] F.J. Kurdahi and A.C. Parker, "REAL: A Program for Register Alloca-
tion", DAC '87, pp. 210-215, 1987.
[3] H.R. Lin and T.T. Hwang. "Power Reduction by Gate Sizing with Path-
oriented Slack Calculation", Proc. of ASP-DAC '95, pp. 7-12, 1995.
[4] Inki Hong and Miodrag Potkonjak, "Power Optimization in Disk-Based
Real-Time Application Specic Systems", ICCAD '96, pp. 634-637,
1996.
[5] V. Tiwari, S. Malik and A. Wolfe, "Compilation Techniques for Low
Energy: an overview", IEEE Symposition on Low Power Electronics,
pp. 38-39, 1994.
[6] Chandrakasan, Sheng and Brodersen, "Low-power technique for
portable real-time DSP applications", Proc. of VLSI Design, 1992.
[7] K. Keutzer "The impact of CAD on the Design of Low Power Digital
Circuits", IEEE Symposium on Low Power Electronics, pp. 42-45, 1994.
[8] Berkelaar, M.R.C.M. and Jess, J.A.G "Gate sizing in MOS digital cir-
cuits with linear programming" EDAC. Proceedings of the European, ,
pp. 217-221, 1990.
[9] V. Tiwari, P. Ashar and S. Malik, "Technology Mapping for Low power",
DAC, , pp. 74-79, June 1993.
[10] B. Lin, "Technology Mapping for Low power Dissipation", VLSI in
Computers and Processors, , Oct. 1993.
[11] A. Chandrakasan, S. Sheng, and R. Broderssen. "Low power CMOS
digital design", IEEE Trans on SSC, , pp. 473-483, Apr 1992.
[12] J. Monteiro, S. Devadas, P. Ashar, and A. Mauskar, "Scheduling Tech-
niques to Enable Power Management", proc. of the 33nd DAC, pp. 349-
352, 1996.
[13] Jui-Ming Chang and Massoud Pedram, "Module Assignment for Low
Power", EDAC 96, 1996.
[14] Anand Raghunathan and Niraj K. Jha, "An Iterative-Improvement-
Based Low-Power Data Path Synthesis System", IEEE TCAD. , Novem-
ber 1997.
[15] Anand Raghunathan, Niraj K. Jha, Sujit Dey, "Power Management in
High-Level Synthesis", IEEE TVLSI. , March, 1999.
[16] J-M. Chang and M. Pedram, "Register Allocation and Binding for Low
Power", Proc. of the 32nd DAC, pp. 29-35, 1995
[17] G. Lakshminarayana, A. Raghunathan, N. K. Jha and S. Dey, "A Power
Management Methodology for High-Level Stnthesis", Proc. Int. Conf.
VLSI Design , pp. 24-29, 1998.
[18] Chien-Jyh Liu and Shi-Yu Huang, "Low-Power Synthesis For Extended
Finite State Machines", Proc. of the 12th VLSI/CAD Symposium, 2001.
[19] E. Musoll and J. Cortadella, "Scheduling and Resource Binding for Low
Power", Proc. of the 8th Int. on System Synthesis , pp. 104-109, 1995.
[20] W. Yang, In-C. Park and C-M Kyung, "Low-Power High-Level Synthesis
Using Latches" ASP-DAC , 2001.
[21] Enoch Hwang, Frank Vahid, Yu-Chin Hsu, "FSMD Functional Parti-
tioning for Low Power" DATE99 , 1999.
[22] L. Benini, P. Vuillod, G. De Micheli and C. Coelho, "Synthesis of Low-
Power Selectively-Clocked Systems from High-Level Specication", In-
ternational Symposium on System Synthesis, pp. 57-63, Nov. 1996.
[23] J. Monteiro and A. Oliveira, "Finite State Machine Decomposition For
Low Power", Proc. of the DAC , pp. 758-763, 1998
[24] M. Alidina, J. Monteiro, S. Devadas, A. Ghosh and M. Pa-
paefthymiou, "Precomputataion-Based Sequential Logic Optimization
for Low Power", IEEE TVLSI , pp 426-436, Dec. 1994.
[25] Vivek Tiwari, Sharad Malik and Pranav Ashar, "Guarded Evaluation:
Pushing Power Management to Logic Synthesis/Design", International
Symposium on Low Power Design, 1995.
[26] F-Y. Chiang and T-T. Hwang, "Resource Allocation for Low Power
VLSI Circuits" thesis paper , 1999.
[27] C. J. Alpert and A. B. Kahng, "Multi-Way Partitioning Via Spacelling
Curves and Dynamic Programing", IEEE DAC94, pp. 652-657, 1994.
[28] S-H. Chow, Y-C. Ho, T-T. Hwang and C. L. Liu, "Low Power Re-
alization of Finite State Machines-A Decomposition Approach", ACM
Transactions on Design Automation of Electronic Systems, vol. 1, no. 3,
pp. 315-340, 1996.
[29] D.D. Gajski. "Principles of digital design", 1997.