由於靜態隨機存取記憶體(SRAM)的快速讀取、寫入速度與將近無限次數的耐寫入、耐讀取特性，使得靜態隨機存取記憶體被廣泛應用於處理器暫存器、高速緩存器及各式各樣的緩衝器中的揮發性記憶體，在記憶體應用領域當中佔有一席之地。此外，隨著通訊系統與物聯網(IoT)技術的快速發展，對於資料傳輸的保密與驗證需求，在近期受到了廣泛的關切。透過特定演算法所產生出的決定性位元亂數(Deterministic Random Bit, DRB)潛在著能被預測的風險，導致資訊傳輸間存有遭受破解並竊取的隱憂。
　　綜觀上述背景，近年來提出了物理不可複製函數(PUF)用以滿足資料通訊的保密與防駭需求。傳統的靜態隨機存取記憶體物理不可複製函數(SRAM PUF)是將二維陣列的靜態隨機存取記憶體作為二維密碼陣列使用，然而其一直存在著輸出響應(Response)的不一致性，導致讀取錯誤的機率，需要使用糾錯編碼算法進行校正。
　　因此，本論文首次提出了一種結合自我修正元件之新型靜態隨機存取記憶體作為物理不可複製函數應用，此新型元件相容於互補式金氧半場效電晶體(CMOS)邏輯製程，不需要加入額外的製程光罩或特殊製程步驟。藉由內嵌自我修正元件在靜態隨機存取記憶體結構下特殊的修正機制，降低製程變異所導致的不一致性，並且利用隨機電報雜訊(Random Telegraph Noise, RTN)的隨機性與不可預測性作為物理不可複製函數所輸入的激勵(Challenge)來源。

Due to the fast read and write speeds of the SRAM and the almost infinite number of write and read resistance characteristic, it’s widely used in processor registers, caches, and various types of buffers for volatile memory. In addition, with the rapid development of communication systems and IoT technologies, the need for confidentiality and verification of data transmission has received widespread concerns. The risk of deterministic bit randomness generated by the algorithm can be predicted. This leads to the potential for information transmission to be compromised and stolen.
　　According to the above background, physical unclonable function has been proposed to satisfy the confidentiality requirements. The conventional SRAM physical unclonable function uses a two-dimensional array of SRAM as a two-dimensional cryptographic array. However, there has always been an inconsistency in the output response, resulting in the probability of reading errors and requiring correction using an error correction coding algorithm.
　　In this work, a novel SRAM combined with self-correcting device as a physical unclonable function application has been firstly proposed. This new device is compatible with CMOS logic processes. We introduce a self-correcting device structure to combine with SRAM cells. Through blanket trimming operation, process variations can be effectively suppressed. Moreover, the randomness and unpredictability of random telegraph noise (RTN) are used as the source of the challenge for the physical unclonable function input.

摘要........................................................ i
Abstract....................................................ii
致謝.......................................................iii
內文目錄....................................................iv
附圖目錄....................................................vi
附表目錄..................................................viii
第一章 序論..................................................1
1.1　　物理不可複製函數應用..............................1
1.2　　論文大綱..........................................2
第二章 物理不可複製函數硬體之發展與文獻回顧..................3
2.1　　光學物理不可複製函數..............................4
2.2　　塗層物理不可複製函數..............................5
2.3　　電阻式隨機存取記憶體物理不可複製函數..............5
2.4　　靜態隨機存取記憶體物理不可複製函數................7
2.5　　小結..............................................9
第三章 結合自我修正元件之新型靜態隨機存取記憶體.............21
3.1　　自我對準氮化矽修正元件介紹.......................22
3.1.1　　 物理不可複製函數修正元件架構..........22
3.1.2　　 製造過程..............................22
3.2　　物理不可複製函數單元操作特性.....................23
3.2.1　　 修正機制..............................23
3.2.2　　 修正電流特性..........................24
3.2.3　　 修正元件電性量測......................24
3.3　　結合自我修正元件之新型靜態隨機存取記憶體.........25
3.4　　小結.............................................25
第四章 新型靜態隨機存取記憶體應用於物理不可複製函數.........35
4.1　　下電路讀取電流的匹配特性.........................35
4.2　　上電路讀取電流的匹配特性.........................36
4.3　　隨機電報雜訊作為物理不可複製函數激勵.............37
4.3.1　　 隨機電報雜訊來源......................37
4.3.2　　 隨機電報雜訊量測結果..................38
4.4　　靜態隨機存取記憶體電路讀取電流的隨機特性.........38
4.5　　唯一性與亂度特性分析.............................38
4.6　　隨機亂度結果分析與驗證...........................39
4.7　　小結.............................................41
第五章 總結.................................................53
參考文獻....................................................54

[1] Gang Li, Pengjun Wang, and Yuejun Zhang, “Highly Reliable Lightweight PUF Circuit with Temperature and Voltage Compensated for Secure Chip Identification,” IEEE 12th International Conference on ASIC (ASICON), 2017.
[2] Dooyoung Kim, M. Adil Ansari, Jihun Jung, and Sungju Park, “SCAN-PUF: PUF Elements Selection Methods for Viable IC Identification,” IEEE 24th Asian Test Symposium (ATS), 2015.
[3] Chongyan Gu, Student Member, IEEE and Maire O’Neill, “Ultra-compact and Robust FPGA-based PUF Identification Generator,” IEEE International Symposium on Circuits and Systems (ISCAS), 2015.
[4] Chongyan Gu, Neil Hanley, and Maire O'Neill, “FPGA-based Strong PUF with Increased Uniqueness and Entropy Properties,” IEEE International Symposium on Circuits and Systems (ISCAS), 2017.
[5] Kota Fruhashi, Mitsuru Shiozaki, Akitaka Fukushima, Takahiko Murayama, and Takeshi Fujino, “The Arbiter-PUF with High Uniqueness utilizing Novel Arbiter Circuit with Delay-Time Measurement,” IEEE International Symposium of Circuits and Systems (ISCAS), 2011.
[6] Takanori Machida, Dai Yamamoto, Mitsugu Iwamoto, and Kazuo Sakiyama, “A New Mode of Operation for Arbiter PUF to Improve Uniqueness on FPGA,” Federated Conference on Computer Science and Information Systems, 2014.
[7] Michael Geis, Karen Gettings, and Michael Vai, “Optical Physical Unclonable Function,” IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS), 2017.
[8] Gabriel C. Birch, Bryana L. Woo, Charles F. LaCasse, Jaclynn J. Stubbs, and Amber L. Dagel, “Computational Optical Physical Unclonable Functions,” International Carnahan Conference on Security Technology (ICCST), 2017.
[9] Boris ˇSkori´c, Geert-Jan Schrijen, Wil Ophey, Rob Wolters, Nynke Verhaegh, and Jan van Geloven, “Experimental Hardware for Coating PUFs and Optical PUFs,” Book of Security with noisy data. Private biometrics, secure storage and anti-counterfeiting (pp.255-268), 2007.
[10] An Chen, “Comprehensive Assessment of RRAM-based PUF for Hardware Security Applications,” IEEE International Electron Devices Meeting (IEDM), 2015.
[11] Mu Li, Peng Huang, Lei Shen, Zheng Zhou, Jin-Feng Kang, and Xiao-Yan Liu, “Simulation of the RRAM-based Flip-Flops with Data Retention,” IEEE International Nanoelectronics Conference (INEC), 2016.
[12] Sreeja Chowdhury, Xiaolin Xu, Mark Tehranipoor, and Domenic Forte, “Aging Resilient RO PUF with Increased Reliability in FPGA,” International Conference on ReConFigurable Computing and FPGAs (ReConFig), 2017.
[13] Supreet Jeloka1, Kaiyuan Yang, Michael Orshansky, Dennis Sylvester, and David Blaauw, “A sequence dependent challenge-response PUF using 28nm SRAM 6T bit cell,” Symposium on VLSI Circuits, 2017.
[14] Hirofumi Shinohara, Baikun Zheng, Yanhao Piao, Bo Liu, and Shiyu Liu, “Analysis and Reduction of SRAM PUF Bit Error Rate,” International Symposium on VLSI Design, Automation and Test (VLSI-DAT), 2017.
[15] Ye Wang and Michael Orshansky, “Efficient Helper Data Reduction in SRAM PUFs via Lossy Compression,” Design, Automation & Test in Europe Conference & Exhibition (DATE), 2018.
[16] Bin Chen, Tanya Ignatenko, Frans M.J. Willems, Roel Maes, Erik van der Sluis, and Georgios Selimis, “A Robust SRAM-PUF Key Generation Scheme Based on Polar Codes,” IEEE Global Communications Conference, 2017.
[17] S. W-C, M. C-Y, C. Y-D, S. S-S, T. M-J, K. Y-C, et al., “Novel Self-Aligned Nitride One Time Programming Memory,” JJAP, vol. 47, p. 8369, 2008.
[18] C. Ying-Je, H. Chia-En, C. Hsin-Ming, L. Han-Chao, J. R. Shih, K. Wu, et al., “A Novel 2-Bit/Cell p-Channel Logic Programmable Cell With Pure 90-nm CMOS Technology,” Electron Device Letters, IEEE, vol. 29, pp. 938-940, 2008.
[19] L. Han-Chao, C. Kai-Yuan, K. Ya-Chin, and L. Chrong-Jung, “A 0.26-μm2 U-Shaped Nitride-Based Programming Cell on Pure 90-nm CMOS Technology,” Electron Device Letters, IEEE, vol. 28, pp. 837-839, 2007.
[20] H. Chia-En, C. Hsin-Ming, L. Han-Chao, C. Ying-Je, K. Ya-Chin, and L. Chrong-Jung, “A New Self-Aligned Nitride MTP Cell with 45nm CMOS Fully Compatible Process,” Electron Devices Meeting, 2007.
[21] H. Chia-En, C. Ying-Je, L. Han-Chao, K. Ya-Chin, and L. Chrong-Jung, “A Study of Self-Aligned Nitride Erasable OTP Cell by 45-nm CMOS Fully Compatible Process,” Electron Devices, IEEE Transactions on, vol. 56, pp. 1228-1234, 2009.
[22] S. Wen Chao, H. Chia-En, O. Hsun, K. Ya-Chin, and L. Chrong-Jung, “32nm strained nitride MTP cell by fully CMOS logic compatible process,” VLSI Technology, Systems, and Applications (VLSI-TSA), 2012.
[23] P. Y. Lin, T. H. Yang, Y. T. Sung, C. J. Lin, Y. C. King, and T. S. Chang, “Self-align nitride based logic NVM in 28nm high-k metalgate CMOS technology,” VLSI Technology, Systems, and Applications (VLSI-TSA), 2013.
[24] Xiaofei Wang, Weichao Xu, and Chris H. Kim, “SRAM Read Performance Degradation under Asymmetric NBTI and PBTI Stress: Characterization Vehicle and Statistical Aging Data,” IEEE Custom Integrated Circuits Conference, 2014.
[25] Sriramkumar Venugopalan, Vivek Joshi, Luis Zamudio, Matthias Goldbach, Gert Burbach, Ralf VanBentum, and Sriram Balasubramanian, “SRAM Read Current Variability and its Dependence on Transistor Statistics,” IEEE Custom Integrated Circuits Conference, 2013.
[26] Joseph Wang, Ping Liu, Yandong Gao, Pankaj Deshmukh, Sam Yang, Ying Chen, Wing Sy, Lixin Ge, Esin Terzioglu, Mohamed Abu-Rahma, Manish Garg, Sei Seung Yoon, Michael Han, Mehdi Sani, and Geoffrey Yeap, “Non-Gaussian Distribution of SRAM Read Current and Design Impact to Low Power Memory using Voltage Acceleration Method,” Symposium on VLSI Circuits, 2017.
[27] Zexuan Zhang, Shaofeng Guo, Xiaobo Jiang, Runsheng Wang, Ru Huangrs, and Jibin Zou, “Investigation on the Amplitude Distribution of Random Telegraph Noise (RTN) in Nanoscale MOS Devices,” IEEE International Nanoelectronics Conference (INEC), 2016.
[28] Giulio Torrente, Niccol`o Castellani, Andrea Ghetti, Christian Monzio Compagnoni, Andrea L. Lacaita, Alessandro S. Spinelli, and Augusto Benvenuti, “Assessment of the Statistical Impedance Field Method for the Analysis of the RTN Amplitude in Nanoscale MOS Devices,” International Conference on Simulation of Semiconductor Processes and Devices (SISPAD), 2013.
[29] Nanbo Gong, Runsheng Wang, Changze Liu, Jibin Zou, and Ru Huang, “On the AC Random Telegraph Noise (RTN) in MOS Devices: An Improved Multi-phonon based Model,” IEEE 11th International Conference on Solid-State and Integrated Circuit Technology, 2012.
[30] D. Veksler, G. Bersuker, B. Chakrabarti1, E. Vogel1, S. Deora, K. Matthews, D. C. Gilmer, H.-F. Li, S.Gausepohl, and P. D. Kirsch, “Methodology for the statistical evaluation of the effect of random telegraph noise (RTN) on RRAM characteristics,” IEEE International Electron Devices Meeting (IEDM), 2012.
[31] A. Sadr and M. Zolfaghari-Nejad, “Weighted Hamming distance for PUF performance evaluation,” Electronics Letters, IET Journals & Magazines, vol. 49, no. 22, pp. 1376–1378, 2013.