由於電阻式隨機存取記憶體元件的小尺寸、快速的寫抹速度、低電壓操作、優異的耐久性以及和CMOS製程的相容性，其電阻開關特性引起了各方諸多興趣。當CMOS的平面製程結構已經達到極限，鉛直的鰭式結構已經確定是用來滿足邏輯晶片的可微縮性以及性能的需求。此外，隨著通信和物聯網（IoT）技術的快速發展，對資訊保密的需求在最近受到了極大的關注。由演算法產生出的偽隨機亂數是可以容易地被預測導致資訊保密降低。由於這個原因，物理不可複製函數(PUF)已經被提出用來滿足這些需求。存在於16nm 鰭式場效電晶體介電質電阻式隨機存取記憶體（FIND RRAM）的隨機電報雜訊（Random Telegraph Noise，RTN）的隨機性和不可預測性初次被應用在與時間連續的物理不可複製函數（PUF）的應用。新的TC-PUF FIND RRAM顯示寬廣的操作範圍在電壓、溫度方面以及透過亂數測試證實了TC-PUF FIND RRAM在頻率，不可預測性和連續性方面的可行性和穩定性。有效率的數據生成速率更讓TC-PUF FIND RRAM在資料安全和辨識應用中成為了一個非常有潛力的解決方案。在這篇論文中，新的TC-PUF可以在每個讀取點生成2D 1kbit真實隨機亂數數據，因此TC-PUF可以沿著讀取的第三維度有效地生成一串隨機和連續的物理不可複製函數（PUF）陣列。此外，產生的數據可以被非揮發性地儲存在其他用於PUF應用的鏡像1kbit FIND RRAM陣列中，而且如果需要更新，數據可以簡單地被下一個時間下取樣的TC-PUF數據取代。

Resistive switching characteristics of the RRAM devices have attracted many interests due to the small cell size, fast P/E speed, low voltage operation, superior endurance, and CMOS process compatibility. As the CMOS process reaches a limit with the planar structure, the vertical FinFET structure has been figured out to serve the need of scalability and performance in logic chips. Besides, with the rapid development of communication and internet of things (IoT) technologies, demand for information security attracts much attention recently. A pseudo random number generator generated by algorithm may be predicted easily to induce the reduction of data security. Physical Unclonable Function (PUF) has been proposed to serve the needs besides RTN is mostly adaptable to the advanced CMOS processes. The randomness and unpredictability of Random Telegraph Noise (RTN) of 16nm FinFET Dielectric (FIND) RRAM is firstly implemented to Time-Contingent Physical Unclonable Function (TC-PUF) application. A novel 3D Time-Contingent Physical Unclonable Function (TC-PUF) realized by 1kbit 16nm FinFET Dielectric (FIND) RRAM has been newly proposed and demonstrated on a pure 16nm FinFET CMOS logic technology. The new TC-PUF FIND RRAM shows wide operation ranges of voltages and temperatures, and Randomness test confirms the feasibility and stability of the TC-PUF FIND RRAM in the frequency, unpredictability, and long-run continuity. The efficient data generation rate further makes the TC-PUF FIND RRAM be a very promising solution in the data security and identification applications. In this project, the novel TC-PUF can generate 2D 1kbit true random number data at each read point, as a result the TC-PUF can efficiently generate a serial random and continuous Physical Unclonable Function (PUF) array along the third dimension of reading time. Besides, the generated data can be non-volatile stored in a mirror 1kbit FIND RRAM array and the data can be easily replaced by the next TC-PUF data sampled to update.

內文目錄
摘要…….. ........................................................................................................ i
Abstract…. ......................................................................................................ii
致謝…….. .....................................................................................................iiv
內文目錄.. ...................................................................................................... vi
附圖目錄.. ................................................................................................... viii
附表目錄.. ....................................................................................................... x
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 .................................. 4
2.1 Optical物理不可複製函數 .......................................................... 5
2.2 Coating物理不可複製函數 ......................................................... 5
2.3 SRAM 物理不可複製函數 .......................................................... 6
2.4 RRAM物理不可複製函數 .......................................................... 7
第三章 鰭式介電質隨機存取電阻式記憶體與隨機電報雜訊 ................ 18
3.1 鰭式場效電晶體介電質隨機存取電阻式記憶體 .................... 18
3.1.1 元件結構與佈局設計 ........................................................ 18
3.1.2 量測環境介紹..................................................................... 19
3.1.3 基本電性 ............................................................................. 19
3.2 隨機電報雜訊 ............................................................................. 20
3.2.1 隨機電報雜訊量測結果 .................................................... 21
3.2.2 電壓對隨機電報雜訊之影響 ............................................ 22
3.2.3 溫度對隨機電報雜訊之影響 ............................................ 22
第四章 三維時間連續物理不可複製函數 ................................................ 37
4.1 三維時間連續不可複製函數概念簡介 ..................................... 37
4.2 記憶胞雜訊之矩陣分佈 ............................................................. 38
4.3 三維時間連續不可複製函數之譜分析 ..................................... 38
4.4 溫度與電壓對三維時間連續不可複製函數特性之影響 ........ 39
4.5 獨特性和亂度特性分析 ............................................................. 39
4.6 三維時間連續不可複製函數電報雜訊數據............................. 40
4.7 隨機亂度結果分析與驗證 ......................................................... 41
4.8 小結 ............................................................................................. 43
第五章 總結及未來展望 ............................................................................. 58
5.1 總結 ............................................................................................. 58
5.2 未來展望 ..................................................................................... 58
參考文獻 60

附圖目錄
圖2.1 物理不可複製函數示意圖 ........................................................... 9
圖2.2 PUF特性示意圖……………………………………………….10
圖2.3 光學物理不可複製函數示意圖 ................................................. 11
圖2.4 Coating PUF架構示意圖 ........................................................... 12
圖2.5 SRAM PUF示意圖 .................................................................... 13
圖2.6 RRAM之電阻分佈圖 ................................................................ 14
圖2.7 (a)RRAM電阻之不規則分佈圖 ............................................... 15
圖2.7 (b)RRAM電阻之尾部彎曲圖 ................................................... 15
圖2.8 不同溫度下PUF轉態示意圖 ................................................... 16
圖2.9 改善RRAM PUF可靠度之方法示意圖 .................................. 17
圖3.1 FIND RRAM結構圖 .................................................................. 23
圖3.2 FIND RRAM之(a) NOR型陣列示意圖。 (b) TEM剖面圖。 ………………………………………………………………….24
圖3.3 FIND RRAM之佈局圖 .............................................................. 25
圖3.4 量測系統示意圖 ......................................................................... 26
圖3.5 FIND RRAM之多次設置、重置直流圖 ................................. 28
圖3.6 FIND RRAM之直流電阻轉換圖 ............................................. 29
圖3.7 FIND RRAM讀取操作時之隨機電報雜訊 ............................. 30
圖3.8 載子流經FIND RRAM介電層被陷阱捕捉、釋放示意圖 .... 31
圖3.9 隨機電報雜訊示意圖 ................................................................. 32
圖3.10 FIND RRAM之LRS與HRS隨機電報雜訊 .......................... 33
圖3.11 不同電壓下之隨機電報雜訊特性 ............................................. 34
圖3.12 不同電壓下之隨機電報雜訊生成頻率 ..................................... 35
圖3.13 不同溫度下之隨機電報雜訊 ..................................................... 36
圖4.1 三維時間連續物理不可複製函數示意圖 ................................. 44
圖4.2 (a)1kbit RTN訊號 ...................................................................... 45
圖4.2 (b) 1kbit數位化訊號 .................................................................. 45
圖4.3 數位化轉換架構圖 ..................................................................... 46
圖4.4 1kbit FIND RRAM電流擾動比率分佈圖 ................................ 47
圖4.5 根據電流擾動比率區分RTN bit和Quiet bit .......................... 48
圖4.6 RTN bit和Quiet bit之功率譜密度分析圖 .............................. 49
圖4.7 (a)電流擾動比率之時間分佈圖 ................................................ 50
圖4.7 (b)RTN bit和Quiet bit之累積機率分佈圖 .............................. 50
圖4.8 (a)不同電壓下之RTN頻率累積機率分佈圖 .......................... 51
圖4.8 (b) 不同溫度下之RTN頻率累積機率分佈圖 ........................ 51
圖4.9 (a) 不同電壓下之電流擾動比率 .............................................. 52
圖4.9 (b)不同溫度下之電流分佈 ........................................................ 52
圖4.10 3D TC-PUF之外部漢明距離 .................................................... 53
圖4.11 3D TC-PUF之WL相依性 ........................................................ 54
圖4.12 (a)3D TC-PUF之一百萬RTN量測.......................................... 55
圖4.12 (b)3D TC-PUF之一百萬數位化輸出 ....................................... 55

附表目錄
表3.1 FIND RRAM之操作電壓 .......................................................... 17
表4.1 PUF特性判斷依據 .................................................................... 56
表4.2 3D TC-PUF之NIST亂數測試 ................................................. 57

[1] Elena Ioana Vatajelu, Giorgio Di Natale, Paolo Prinetto, Politecnico di Torino, “Security Primitives (PUF and TRNG) with STT-MRAM,” 34th VLSI Test Symposium, IEEE, 2016.
[2] Siam U. Hussain, Mehrdad Majzoobi, and Farinaz Koushanfar, “A Built-in-Self-Test Scheme for Online Evaluation of Physical Unclonable Functions and True Random Number Generators,“ IEEE Transactions On Multi-scale Computing Systems, 2016.
[3] Masoud Rostami, James B. Wendt, Miodrag Potkonjak, and Farinaz Koushanfar, “Quo Vadis, PUF Trends and Challenges of Emerging Physical-Disorder based Security,” EDAA, 2014.
[4] Ilia A, Bautista Adames, Jayita Das, and Sanjukta Bhanjar, “Survey of emerging technology based physical unclonable functions,” GLSVLSI, 2016, pp. 317-322.
[5] Yijun Cui, Chenghua Wang, Weiqiang Liu, Yifei Yu, Maire O’Neill, and Fabrizio, “Low-cost configurable ring oscillator PUF with improved uniqueness,” ISCAS, 2016, pp. 558-561.
[6] Chayanika Roy Chaudhuri, Fathi Amsaad, and Mohammed Niamat, “Impact of temporal variations on the performance and reliability of configurable ring oscillator PUF,” IEEE, 2016.
[7] Roel Maes and Ingrid Verbauwhede, “Physically Unclonable Functions: a Study on the State of the Art and Future Research Directions,” : https://www.researchgate.net/publication/226371108
[8] L. Torres, R.M. Brum,L.V. Cargnini, G. Sassatelli, “Trends on the application of emerging nonvolatile memory to processors and programmable devices,” IEEE International Symposium on Circuits and Systems, pp.101-104, 2013.
[9] An Chen, “Comprehensive assessment of RRAM-based PUF for hardware security applications,” IEEE International Electron Devices Meeting(IEDM), pp.10.7.1-10.7.4, 2015.
[10] Hsin Wei Pan, Kai Ping Huang, Shih Yu Chen, Ping Chun Peng, Zhi Sung Yang, Cheng-Hsiung Kuo, Yue-Der Chih, Ya-Chin King, and Chrong Jung Lin, “1Kbit FinFET Dielectric (FIND) RRAM in pure 16nm FinFET CMOS logic process,” IEEE International Electron Devices Meeting(IEDM), pp.10.5.1-10.5.4, 2015.
[11] Yuan Heng Tseng, Wen Chao Shen, Chia-En Huang, Chrong Jung Lin,
61
Ya-Chin King, “Electron Trapping Effect on the Switching Behavior of Contact RRAM Devices through Random Telegraph Noise Analysis,“ IEDM10-639, 2010.
[12] Wei Feng, Chun Meng Dou, Masaaki Niwa, Keisaku Yamada, and Kenji Ohmori, ”Impact of Random Telegraph Noise Profiles on Drain-Current Fluctuation During Dynamic Gate Bias,” IEEE EDL, 2014.
[13] Lee, Jung-Kyu, Ju-Wan Lee, Jinwon Park, Sung-Woong Chung, Jae Sung Roh, Sung-Joo Hong, Il-whan Cho, Hyuck-In Kwon, and Jong-Ho Lee, ”Extraction of trap location and energy from random telegraph noise in amorphous TiOx resistance random access memories,” Applied Physics Letters 98, 2011.
[14] W. Feng, K. Yamada, and K. Ohmori, “Random Telegraph Noise Induced Drain-Current Fluctuation during Dynamic Gate Bias in Si MOSFETs,” IEEE, 2013.
[15] C. Mukherjee, T.jacquet, T.Zimmer, C.Maneux, A.Chakravorty, J.Boeck, and K.Aufinger, “Comprehensive study of random telegraph noise in base and collector of advanced SiGe HBT: Bias, geometry and trap locations,” ESSDERC, 2016, pp.260-263.
[16] M. V. Haartman, M. Sanden, and M. Ostling, “Random telegraph signal noise in SiGe heterojunction bipolar transistors”, J. Appl. Phys., VOL. 92, NO. 8 pp. 4414-4421, 2002.
[17] Maurício Banaszeski da Silva, Hans P. Tuinhout, Adrie Zegers-van Duijnhoven, Gilson I. Wirth, and Andries J. Scholten, “A Physics-Based Statistical RTN Model for the Low Frequency Noise in MOSFETs,“ IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 63, NO. 9, SEPTEMBER 2016.
[18] J. S. Kolhatkar, L. K. J. Vandamme, C. Salm, and H. Wallinga, “Separation of random telegraph signals from 1/f noise in MOSFETs under constant and switched bias conditions,” ESSDERC, pp. 549–552, 2003.
[19] Yuki Furubayashi, Matsuto Ogawa, and Satofumi Souma, “Numerical simulation of current noise caused by potential fluctuation in nanowire FET with an oxide trap,” IEEE, 2014.
[20] Eduardo Pérez, Alessandro Grossi, Cristian Zambelli, Piero Olivo, and Christian Wenger, ”Impact of the Incremental Programming Algorithm on the Filament Conduction in HfO2-Based RRAM Arrays,” IEEE Journal of the Electron Devices Society, 2017.
[21] Kyosuke Ito, Takashi Matsumoto, Shinichi Nishizawa, Hiroki
62
Sunagawa,Kazutoshi Kobayashi and Hidetoshi Onodera, ”Modeling of Random Telegraph Noise under Circuit Operation-Simulation and Measurement of RTN-induced delay fluctuation,” IEEE, 2011.
[22] Charles Surya, Sze-Him Ng, Elliott R. Brown, and Paul A. Maki, ”Spectral and Random Telegraph Noise Characterizations of Low-Frequency Fluctuations in GaAs/Al0.4Ga0.6As Resonant Tunneling Diodes,” IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 41, NO. 11, NOVEMBER 1994.
[23] Yusuke Higashi, Nobuyuki Momo, Hiroki Sasaki, Hisayo Sasaki Momose, Tatsuya Ohguro, Yuichiro Mitani, Takamitsu Ishihara, and Kazuya Matsuzawa, ”Unified Transient and Frequency Domain Noise Simulation for Random Telegraph Noise and Flicker Noise Using a Physics-Based Model,” IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 61, NO. 12, DECEMBER 2014.
[24] Min-Kyu Jeong, Sung-Min Joe, Chang-Su Seo, Kyung-Rok Han, Eunseok Choi, Sung-Kye Park, and Jong-Ho Lee, ”Analysis of Random Telegraph Noise and Low Frequency Noise Properties in 3-D Stacked NAND Flash Memory with Tube-Type Poly-Si Channel Structure,” VLSI Technology (VLSIT), 2012.
[25] M.Maestro, A. Crespo-Yepes, J. Martin-Martinez, S. Claramunt, R. Rodriguez, M. Nafria, X. Aymerich, ”Distinguishing conductive filament and non-localized gate conduction in resistive switching devices,” 10th Spanish Conference on Electron Devices (CDE), 2015.
[26] Xiaonan Yang, Jing Liu, Zhiwei Zheng, Yan Wang, Dandan Jiang, Shengfen Chiu, Hanming Wu, Ming Liu, “Impact of P/E Cycling on Read Current Fluctuation of NOR Flash Memory Cell: A Microscopic Perspective Based on Low Frequency Noise Analysis,” IEEE International Reliability Physics Symposium, 2015.
[27] D. Veksler, G. Bersuker, B. Chakrabarti1, E. Vogel1, S. Deora, K. Matthews, D. C. Gilmer, H.-F. Li, S.Gausepohl, 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.
[28] T. Nagumo, K. Takeuchi, S. Yokogawa, K. Imai, and Y. Hayashi, “New Analysis Methods for Comprehensive Understanding of Random Telegraph Noise,” IEEE International Electron Devices Meeting (IEDM), 2009.
[29] Jiezhi Chen, Tetsufumi Tanamoto, Hiroki Noguchi and Yuichiro Mitani,
63
“Further Investigations on Traps Stabilities in Random Telegraph Signal Noise and the Application to a Novel Concept Physical Unclonable Function (PUF) with Robust Reliabilities,” VLSI Technology Digest of Technical Papers, 2015.
[30] National institute of standards and technology, “A statistical test suite for random and pseudorandom number generators for cryptographic applications,” Pub 800-22, 2001.
[31] Charles Herder, Meng-Day (Mandel) Yu, Farinaz Koushanfar, and Srinivas Devadas, “Physical unclonable functions and applications: a tutorial,” JPROC, pp.1126-1141, 2014.