近似計算為針對可容忍錯誤之應用的新興設計範例，這類的應用如訊數處理、機器學習.......等。在近似計算中，可以通過換取電路的準確性來改善近似電路的面積、延遲或功耗。我們將於此論文提出一種基於節點合併技術並具有錯誤率保證的近似邏輯合成方法。提出之方法的想法是用定值替換電路中的點，並合併電路中功能相似的兩個節點。我們在IWLS2005和MCNC的電路上進行實驗，實驗結果顯示我們提出的方法能夠減少多達80%的面積，平均減少31%的面積。與最新的方法相比，我們提出的方法在相同的5%錯誤率限制下具有51倍的加速。

Approximate computing is an emerging design paradigm for error-tolerant applications. e.g., signal processing and machine learning. In approximate computing, the area, delay, or power consumption of an approximate circuit can be improved by trading off its accuracy. In this paper, we propose an approximate logic synthesis approach based on a node-merging technique with an error rate guarantee. The ideas of our approach are to replace internal nodes by constant values and to merge two similar nodes in the circuit in terms of functionality. We conduct experiments on a set of IWLS 2005 and MCNC benchmarks. The experimental results show that our approach can reduce area by up to 80%, and 31% on average. Compared with the state-of-the-art method, our approach has a speedup of 51 under the same 5% error rate constraint.

中文摘要 i
Abstract ii
誌謝辭 iii
Contents iv
List of Tables vi
List of Figures vii
1 Introduction 1
2 Preliminaries 3
2.1 Error Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.3 Node-Merging Approach . . . . . . . . . . . . . . . . . . . . . . . . . 4
3 Proposed Approach 7
3.1 Node to Constant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.2 Node Replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.3 Error Rate Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.4 Overall Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4 Experimental Results 16
4.1 Probability p in Node to Constant Phase . . . . . . . . . . . . . . . . 16
4.2 Circuit Size Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5 Conclusion 21
6 Acknowledgement 22

1] L. Chakrapani et al., “A Probabilistic Boolean Logic for Energy Efficient Circuit and System Design,” Proc. ASP-DAC, 2010, pp. 628-635.
[2] Y.-C. Chen et al., “Fast Detection of Node Mergers Using Logic Implications,” Proc. ICCAD, 2009, pp. 785-788.
[3] Y.-C. Chen et al., “Fast Node Merging with Don’t Cares Using Logic Implications,” IEEE TCAD, 2010, pp. 1827-1832.
[4] V. Chippa et al., “Analysis and Characterization of Inherent Application Resilience for Approximate Computing,” Proc. DAC, 2013, pp. 1-9.
[5] V. Chippa et al., “Dynamic Effort Scaling: Managing the Quality-Efficiency Tradeoff,” Proc. DAC, 2011, pp. 603-608.
[6] V. Gupta et al., “IMPACT: IMPrecise adders for low-power Approximate CompuTing,” Proc. ISLPED, 2011, pp. 409-414.
[7] J.Han et al., “Approximate Computing: An Emerging Paradigm for EnergyEfficient Design,” Proc. ETS, 2013, pp. 1-6.
[8] K. He, et al., “Controlled Timing-Error Acceptance for Low Energy IDCT Design,” Proc. DATE, 2011, pp. 1-6.
[9] M. Imani et al., “Resistive Configurable Associative Memory for Approximate Computing,” Proc. DATE, 2016, pp. 1327-1332.
[10] Y. Kim et al., “An Energy Efficient Approximate Adder with Carry Skip for Error Resilient Neuromorphic VLSI Systems,” Proc. ICCAD, 2013, pp. 130-137.
[11] T. Kirkland et al., “A Topological Search Algorithm for ATPG,” Proc. DAC, 1987, pp. 502-508.
[12] P. Kulkarni et al., “Trading Accuracy for Power with an Underdesigned Multiplier Architecture,” Proc. VLSID, 2011, pp. 346-351.
[13] K. Y. Kyaw et al., “Low-Power High-Speed Multiplier for Error-Tolerant Application,” Proc. EDSSC, 2010, pp. 1-4.
[14] Y.-A Lai et al., “Efficient Synthesis of Approximate Threshold Logic Circuits with an Error Rate Guarantee,” Proc. DATE, 2018, pp. 773-778.
[15] J. Miao et al., “Approximate Logic Synthesis under General Error Magnitude and Frequency Constraints,” Proc. ICCAD, 2013, pp. 779-786.
[16] A. Mishchenko, “A Catalog of Three-Variable Or-Invert and And-Invert Logical Circuits,” IEEE Trans. Electronic Computers, 1963, pp. 198-223.
[17] N. Shanbhag, “Reliable and Energy-Efficient Digital Signal Processing,” Proc. DAC, 2002, pp. 830-835.
[18] S. Su et al., “Efficient Batch Statistical Error Estimation for Iterative Multilevel Approximate Logic Synthesis,” Proc. DAC, 2018, pp. 1-6.
[19] S. Venkataramani et al., “Substitute-and-Simplify: A Unified Design Paradigm for Approximate and Quality Configurable Circuits,” Proc. DATE, 2013, pp. 1367-1372.
[20] Y. Wu et al., “An Efficient Method for Multi-level Approximate Logic Synthesis under Error Rate Constraint,” Proc. DAC, 2016, pp. 1-6.
[21] S. Yang, “Logic Synthesis and Optimization Benchmarks,” Microelectronics Center of North Carolia, Tech. Rep., 1991.
[22] Y. Yao et al., “Approximate Disjoint Bi-decomposition and Its Application to Approximate Logic Synthesis,” Proc. ICCD, 2017, pp. 517-524.
[23] N. Zhu et al., “An Enhanced Low-Power High-Speed Adder for Error-Tolerant Application,” Proc. ISIC, 2009, pp. 69-72.
[24] Berkeley Logic Synthesis and Verification Group. ABC: A System for Sequential Synthesis and Verification [Online]. Available: http://www.eecs.berkeley.edu/~alanmi/abc
[25] http://iwls.org/iwls2005/benchmarks.html