近年來，智慧型手機的便攜性、便利性使得智慧型手機成為普及的行動裝置，隨著智慧型手機功能導向的發展，被廣泛應用於通話、交易、付費以及身分認證，若遭盜用損失難以估計。上述許多功能需使用者做動態動作將手機與機器或與人溝通，這類應用程式目前僅能夠在使用前透過密碼驗證方式識別，若使用前皆需登入驗證則使用者很可能會為了方便使用而紀錄密碼，導致登入驗證形同虛設。因此本論文目的是要利用智慧型手機內建之Accelerometer及Orientation sensor達到非侵入式(non-intrusive)識別使用者動態動作的行為，以提高智慧型手機的安全等級。本論文用手機最常用之通話行為驗證動態動作是否可作為識別使用者，實驗結果同等錯誤率(Equal Error Rate)為15.4%，未達到低於12.8%目標值，認為動態動作之行為作為識別機制仍需加強其安全性。

In recent years, the smart phones have the portable and convenient ability. They become the most popular mobile devices. The functionality of smart phone development has been enhanced more than before. It has been used for communication, transaction, bill payment, and ID identification. If someone illegally uses your smart phone will make you suffer heavy losses. Most of all the above mentioned functionalities are needed to dynamic motion the smart phone communicate with machine or people. The application in current situation is just only using password authentication. If every application needs this authentication method will make user save password avoid inconvenience, so that authentication becomes useless. So the purpose of this paper use the smart phone’s embedded both accelerometer sensor and orientation sensor, achieve the non-intrusive dynamic motion authentication, and enhance smart phone security level. We use the most common communication behavior one of dynamic motion, if it can be used to verify user himself or not. The equal error rate (EER) of experimental result is under 15.4% not reach the goal, EER below 12.8%. The dynamic motion authentication needs to enhance the security level.

中文摘要............................................................ i
Abstract........................................................... ii
誌謝.............................................................. iii
目錄............................................................... iv
圖目錄............................................................. vi
表目錄........................................................... viii
第一章 序論...................................................... 1
1-1 前言...................................................... 1
1-2 研究動機.................................................. 2
1-3 研究目的.................................................. 4
1-4 論文架構.................................................. 5
第二章 文獻探討.................................................. 6
第三章 實驗設計.................................................. 9
3-1 資料收集.................................................. 9
3-1-1 資料收集環境........................................ 9
3-1-2 資料收集情境....................................... 10
3-2 計算行為特徵............................................. 11
3-3 資料前置處理............................................. 13
3-4 軌跡資料分析識別......................................... 13
3-4-1 抽樣(sampling)..................................... 14
3-4-2 單類分類器(one-class classifier)................... 16
3-4-3 模型訓練與測詴..................................... 18
第四章 實驗結果與分析........................................... 19
4-1 實驗結果................................................. 19
4-2 實驗數據分析............................................. 21
4-3 討論：移動路徑特徵....................................... 24
第五章 研究討論與未來展望....................................... 27
5-1 結論..................................................... 27
5-2 未來展望................................................. 27
參考文獻........................................................... 29
附錄一 特徵計算公式............................................. 31
附錄二 各threshold value實驗數據............................... 32
附錄三 收集之各段軌跡個別建模調整threshold結果................. 34
附錄四 實驗數據總表............................................. 36
附錄五 移動路徑實驗數據總表..................................... 38
附錄六 使用兩種sensor與單一sensor實驗討論..................... 40

[1] Ashbourn, J., “Biometric. Advanced Identity Verification. The Complete Guide” Springer, Berlin Heidelberg New York, 2000.
[2] Clarke, N., Furnell, S., Rodwell, P.and Reynolds, P. “Acceptance of subscriber authentication for mobile telephony devices”, Computers & Security, Vol 21(3), pp. 220-228, 2002.
[3] David M.J. Tax, “Data description toolbox (dd_tools) toolbox version 1.7.3.”, 2009, available from: http://homepage.tudelft.nl/n9d04/dd_tools.html
[4] Defeng Wang, Daniel S. Yeung and Eric C. C. Tsang, “Structured One-Class Classification” IEEE Transactions on Systems, Man and Cybernetics, Part B, vol. 36, no. 6, pp. 1283–1295, 2006.
[5] D.M.J. Tax, “One-class classification; Concept-learning in the absence of counter-examples”, Ph.D. thesis Delft University of Technology, ASCI Dissertation Series, 65, Delft, 19 June 2001, 1-190.
[6] E. Rukzio. “Physical Mobile Interactions: Mobile Devices as Pervasive Mediators for Interactions with the Real World.” PhD thesis, LMU Munchen, 2006.
[7] Gartner, Inc. “Gartner Identifies the Top 10 Consumer Mobile Applications for 2012”, STAMFORD, Conn., 18 November 2009, available from: http://www.gartner.com/it/page.jsp?id=1230413
[8] Gartner, Inc. “Gartner Says Worldwide Mobile Device Sales to End Users Reached 1.6 Billion Units in 2010; Smartphone Sales Grew 72 Percent in 2010”, Egham, UK,9 February 2011, available from: http://www.gartner.com/it/page.jsp?id=1543014
30
[9] Google, Android? Platform, available from: http://developer.android.com/index.html
[10] Khan, S. S., ''Kernels for One-Class Nearest Neighbour Classification and Comparison of Chemical Spectral Data'', Unpublished master''s thesis, National University of Ireland Galway, Galway, Ireland, 2010.
[11] N.L. Clarke and S.M. Furnell, “Advanced user authentication for mobile devices“, Computers & Security, Vol 26, pp. 109-119, 2007.
[12] N. L. Clarke and S. M. Furnell, “Authenticating mobile phone users using keystroke analysis”, International Journal of Information Security, Vol 6(1), pp.1-14, 2007
[13] S. Fuenell, “Beyound the PIN: Enhancing user authentication for mobile devices”, Computer Fraud & Security, pp. 12-17, Aug. 2008.