臺灣博碩士論文加值系統

次世代網路的設計是著眼於未來需求的一種通訊基礎架構。NGN技術特點是，以“IP融合”的網路架構，即在網際網路上的IP技術應用於次世代網路通訊。作為一種通訊基礎架構，次世代網路必需提供使用者在可靠性、耐久性以及服務品質等特性來達成“電信等級”的網路，同時提供易於使用的創新服務。網路效能的關鍵在於精確的時間和頻率。因此，次世代網路的底層技術必需要有更精確的時間和頻率。
另一方面，移動設備是次世代網路的終端。智慧型手機是個人通訊系統的用戶端，它具有與傳統手機、掌上型電腦和桌上型電腦之整合功能，且適用於啇業應用、娛樂、行動通訊和網路功能的設備。若想要讓使用者享用更多元的服務，設計者就必需內建更高階的人機互動應用程序在行動通訊設備中。
一個時間序列是指在連續且均勻時間點上測量所得的一組資料數列。時間序列分析，包括分析時間序列數據的方法，採掘有意義的統數據及其他特徵。時間序列預測模型之基礎，是一種利用已知的過去事件預測未來，以之前蒐集來的資料作為預測的技術。
本論文將提出幾個新穎的、有效能的、穩定的以及具方便性的方法，使用計算智慧於時間序列的訊號處理上，應用在行動通訊網路之頻率同步及手機上的使用者互動介面等領域。
在這個研究中，第二章，提出一種新的方法來解決頻率校正。設計一個控制系統，毎隔2秒蒐集控制信號，Fuzzy以及ANFIS控制器使用這些時間序列來預測新的控制訊號，進而控制從動裝置。第三和第四章，提出2種新的方法來解決智慧型手機上的三維手寫手勢識別。當使用者握住手機揮動，產生三維手寫手勢時，加速度感測器蒐集加速度資料時間序列，使用多組手寫手勢來訓練辯識模組，作為手寫手勢辯識使用。

The next generation network (NGN) is a communication infrastructure designed to address the needs of the coming age. A technical feature of the NGN is that it takes “IP convergence” network architecture, meaning that IP technology developed on the Internet is applied to the NGN. As a communication infrastructure, the NGN needs to provide carrier-grade qualities in terms of reliability, durability, and quality of service (QoS), while providing ease of new service creation. Time and frequency accuracy is critical to the efficiency of network. Thus, underling technologies of NGN needs more precise and accuracy on timing and frequency.
On the other hand, the mobile device is a terminal of NGN. Smart phone is a personal digital client, which is set with those usual functions of traditional mobile phone, PDA and computer and is mixed with features of business, entertainment, mobile and network. The more application we want to use, the more interact between human and device we need to build mobile devices in.
A time series is a sequence of data points, measured typically at successive times spaced at uniform time intervals. Time series analysis comprises methods for analyzing time series data in order to extract meaningful statistics and other characteristics of the data. Time series forecasting is the use of a model to forecast future events based on known past events, to predict data points before they are measured.
This dissertation addresses on some signal processing of time series using the methods of computational intelligence to finding a novel , effective, stable and convenient ways in network of frequency calibration and mobile device of user interface areas.
In this work, chapter 2, a novel method is proposed to solving frequency calibration. The control signal is collected every two second for getting a time series sequences in this control system. Fuzzy and ANFIS controller using those sequences to control slave device. Chapter 3 and 4, two novel methods are proposed to solving 3D handwriting gesture recognition on a smart phone. The mobile device collects accelerations using accelerometer when a user does a gesture with holding the device. The accelerations is a set of time series sequences. To train those sets to get some models for gesture recognition.

List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Chapter 1: Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Scope of the work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Dissertation Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Chapter 2: Computational Intelligence . . . . . . . . . . . . . . . . . . . . . . . . 5
2.1 Fuzzy Control System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Adaptive Neural-Fuzzy Inference System (ANFIS) . . . . . . . . . . . . . . . 6
2.3 Support vector machines(SVM) . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.4 Hidden Markov model (HMM) . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.5 Gaussian mixture model (GMM) . . . . . . . . . . . . . . . . . . . . . . . . . 11
Chapter 3: Frequency Calibration based on the ANFIS . . . . . . . . . . . . . . . 12
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2 System architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.3 The proposed method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.3.1 Fuzzy Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.3.2 ANFIS architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.4 Experimental results and analysis . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Chapter 4: WLCS for 3D Handwriting Recognition on Handheld Devices . . . . . 23
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.2 Theoretical Backgrounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.2.1 The Weighted LCS Algorithm . . . . . . . . . . . . . . . . . . . . . . . 24
4.3 The Feasibility of Applying Weighted LCS in 3D Handwriting Recognition . . 27
4.3.1 Data Sets and Data Preprocessing . . . . . . . . . . . . . . . . . . . . 27
4.3.2 Performance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
. . . . . . . . . . 11
Chapter 3: Frequency Calibration based on the ANFIS . . . . . . . . . . . . . . . 12
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2 System architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.3 The proposed method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.3.1 Fuzzy Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.3.2 ANFIS architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.4 Experimental results and analysis . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Chapter 4: WLCS for 3D Handwriting Recognition on Handheld Devices . . . . . 23
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.2 Theoretical Backgrounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.2.1 The Weighted LCS Algorithm . . . . . . . . . . . . . . . . . . . . . . . 24
4.3 The Feasibility of Applying Weighted LCS in 3D Handwriting Recognition . . 27
4.3.1 Data Sets and Data Preprocessing . . . . . . . . . . . . . . . . . . . . 27
4.3.2 Performance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
i
4.3.3 Experimental Setup and Results . . . . . . . . . . . . . . . . . . . . . 30
4.4 The Proposed 3D Handwriting Recognition System . . . . . . . . . . . . . . . 35
4.4.1 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.4.2 Data Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.4.3 Data Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.4.4 Pattern Recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.5 Experimental Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Chapter 5: GMM for 3D Handwriting Recognition on Handheld Devices . . . . . 43
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
5.2 The typical exampe of HMM . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
5.3 The Proposed 3D Handwriting Recognition System . . . . . . . . . . . . . . . 47
5.3.1 Data Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
5.3.2 Pattern Recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.3.3 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.4 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Chapter 6: Conclusion and Future work . . . . . . . . . . . . . . . . . . . . . . . . 54
6.1 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
6.2 Future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
6.2.1 Calibration of Time and Frequency . . . . . . . . . . . . . . . . . . . . 54
6.2.2 3D Handwriting Recognition . . . . . . . . . . . . . . . . . . . . . . . 55
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

[3GPP, 2005] 3GPP (2005). Requirements for support of radio resource management (fdd).
3GPP Release 6, V 6.12.0.
[Amir and Shalom, 2009] Amir, A. Z. G. and Shalom, B. R. (2009). Weighted lcs. IWOCA
2009, LNCS 5874, pages 36 – 47.
[Baek and Yun, 2008] Baek, J. and Yun, B. J. (2008). A sequence-action recognition applying
state machine for user interface. IEEE Transactions on Consumer Electronics,
54(2).
[Barbancho et al., 2007] Barbancho, J., Leon, C., Molina, F., and Barbancho, A. (2007).
Using artificial intelligence in routing schemes for wireless networks. Computer Com-
munications, 30(14-15):2802 – 2811. Data-aggregation;MAC protocols;Network algorithms;
OLIMPO;Routing paradigms;Wireless sensor nodes;.
[Baum, 1972] Baum, L. E. (1972). An inequality and associated maximization technique in
statistical estimation for probabilistic functions of markov processes. Inequalities, 3:1 –
8.
[Baum and Egon, 1967] Baum, L. E. and Egon, J. A. (1967). An inequality with applications
to statistical estimation for probabilistic functions of a markov process and to a
model for ecology. Bull. Amer. Meteorol. Soc., 73:360 – 363.
[Baum and Petrie, 1966] Baum, L. E. and Petrie, T. (1966). Statistical inference for probabilistic
functions of finite state markov chains. Ann. Math. Statist., 37(6):1554 – 1563.
[Baum et al., 1970] Baum, L. E., Petrie, T., and Soules, G. (1970). A maximization technique
occurring in the statistical analysis of probabilistic functions of markov chains.
Ann. Math. Stat., 41:164 – 171.
[Baum and Sell, 1967] Baum, L. E. and Sell, G. R. (1967). Growth functions for transformations
on manifolds. Pac. J. Math, 27(2):211 – 227.
[Chen and Petriu, 2008] Chen, Qing, N. D. G. and Petriu, E. M. (2008). Hand gesture
recognition using haar-like features and a stochastic context-free grammar. IEEE Trans.
on Instrumentation and Measurement, 57(8):1562 – 1571.
[Cho and Choi, 2006] Cho, S. J. and Choi, E. a. (2006). Two-stage recognition of raw
acceleration signals for 3-d gesture- understanding cell phones.
[Choi and Bang, 2006] Choi, E.-S. and Bang, W.-C. a. (2006). Beatbox music phone: Gesture
interactive cell phone using tri-axis accelerometer.
[Choi et al., 2006] Choi, S.-D., Lee, A., and Lee, S.-Y. (2006). On-line handwritten character
recognition with 3d accelerometer. pages 845–850.
[Cormen and Rivest, 1985] Cormen, T. H., C. E. L. and Rivest, R. L. (1985). Introduction
to Algorithms. The MIT Press.
[Cortes and Vapnik, 1995] Cortes, C. and Vapnik, V. (1995). Support-vector networks.
20:273 – 297.
[D. Drianko, 1992] D. Drianko, H. Hellendoorn, M. R. (1992). An Introduction to Fuzzy
Control. Springer-Verlag.
[Davis and Rougeaux, 1999] Davis, J. A. and Rougeaux, B. (1999). Development of a computer
model of a gps disciplined oscillator to aid error budget determination. volume 1,
pages 291 – 295, United States.
[Dubois D, 1998] Dubois D, P. H. (1998). An introduction to fuzzy systems. Clin. Chim.
Acta., (270):3 – 29.
[Eskelinen, 1999] Eskelinen, P. (1999). Observations on stability measurements of commercial
atomic clocks. Proceedings of the Annual IEEE International Frequency Control
Symposium, 1:186 – 189.
[Fawcett, 2006] Fawcett, T. (2006). An introduction to roc analysis. pages 861 – 874.
[Fraser, 1992] Fraser, J. (1992). What is artificial intelligence? Petroleum Review,
46(541):66, 68 – 69. Artificial intelligence disciplines;Knowledge-based systems;Natural
language;.
[Guler I, ] Guler I, U. E. Application of adaptive neuro-fuzzy inference system for detection
of electrocardiographic changes in patients with partial epilepsy using feature extraction.
Expert Syst. Appl., 33.
[IEEE, 2005] IEEE (2005). Ieee std. 802.16e-2005 and ieee 802.16-2004/cor1-2005. IEEE,
New York, USA.
[Isasi et al., 2007] Isasi, P., Quintana, D., Saez, Y., and Mochon, A. (2007). Applied computational
intelligence for finance and economics. Computational Intelligence, 23(2):111
– 116. Economics problems;Finance problems.
[Jang, 1993] Jang, J.-S. (1993). Anfis: adaptive-network-based fuzzy inference system.
IEEE Transactions on Systems, Man and Cybernetics, 23(3):665 – 685.
[Miller AS, 1990] Miller AS, Blott BH, H. T. (1990). Review of neural network applications
in medical imaging and signal processing. Med. Biol. Eng. Comput., (30):449 – 464.
[Mitra and Acharya, 2007] Mitra, S. and Acharya, T. (2007). Gesture recognition: a survey.
IEEE Trans. on Sys., Man, and Cyber, (3):311 – 324.
[Naso and Maione, 2001] Naso, D. and Maione, G. (2001). Recent developments in the application
of computational intelligence to multi-agent manufacturing control. volume 2,
pages 990 – 994, Melbourne, Australia. Fuzzy controller;Multi-agent manufacturing control;.
[Nicholls and Carleton, 2004] Nicholls, C. and Carleton, G. (2004). Adaptive ocxo drift
correction algorithm. pages 509 – 517, Montreal, Canada.
[R., 1989] R., R. L. (1989). Tutorial on hidden markov models and selected applications in
speech recognition. Proceedings of the IEEE, 77(2):257 – 286.
[Rabiner and Juang, 1986] Rabiner, L. R. and Juang, B.-H. (1986). Introduction to hidden
markov models. IEEE ASSP magazine, 3(1):4 – 16.
[S. Kallio and J.Mantyjarvi, 2003] S. Kallio, J. K. and J.Mantyjarvi (2003). Online gesture
recognition system for mobile interaction. pages 2070 – 2076.
[TTA, 2004] TTA (2004). Specifications for 2.3 ghz band portable internet servicevphysical
layer. TTA, Korea.
[Tu et al., 2008] Tu, K.-Y., Hsu, W.-H., Wu, J.-S., and Liao, C.-S. (2008). Frequency
calibration based on adaptive neural-fuzzy inference system. pages 616 – 617, Broomfield,
CO, United states.
[Vapnik, 1999] Vapnik, V. (1999). The Nature of Statistical Learning Theory. Springer
Press, New York.
[Vasilakos and Spyrou, 2008] Vasilakos, A. V. and Spyrou, G. (2008). Computational intelligence
in medicine and biology: A survey. Journal of Computational and Theoretical
Nanoscience, 5(12):2365 – 2376. Computational intelligence;Ethics;Genomics;Machine
learning;Medical informatics;Proteomics;.
[Viterbi, 1995] Viterbi, A. J. (1995). Principles of Spread Spectrum Communication. Addison
Wesley.
[Wagner and Fischer, 1974] Wagner, R. and Fischer, M. (1974). The string-to-string correction
problem. J. ACM, 21:168 – 173.
[WG., 1990] WG., B. (1990). Use of an artificial neural network for data analysis in clinical
decision making: the diagnosis of acute coronary occlusion. Neural Comput., (2):480 –
489.
[Zhang and Lin, 2002] Zhang, Y.-Q. and Lin, T. (2002). Computational web intelligence
(cwi): Synergy of computational intelligence and web technology. volume 2, pages 1104
– 1107, Honolulu, HI, United states. Computational web intelligence;Electronic business;
Quality of intelligence;Web technology;Wireless networks.