臺灣博碩士論文加值系統

在本篇論文裡，我們將討論分析一個檢測理論的議題：如何在失去資料後重新做決策。考慮一個統計假設檢定，一開始經由某些收集到的資訊做出一個二位元的最佳化判斷，但後來部分的資訊不幸地丟失了，換句話說，只剩下倖存的數據與先前的二位元決定，此時我們需要重新取得決策。該怎麼做才是最好的決定？失去資料後的重新決策這類的問題在日常生活中顯而易見，所以容易被廣泛的應用。論文中提出了一個新穎的檢測方式，利用 Dempster-Shafer (D-S) 理論，又稱證據理論，來解決上述的問題。
證據理論類似人類的決策邏輯，能夠綜合考慮多方資訊且具有處理"不確定"訊息的特性；透過證據理論，我們可以設定對證據的信任度並結合證據來做出決策，同時我們也利用權重的分配來提高決策性能。

中文摘要 i
ABSTRACT ii
誌謝 iii
CONTENTS iv
LIST OF FIGURES vi
LIST OF TABLES viii
Chapter 1 Introduction 1
Chapter 2 Background Knowledge 4
2.1 Neyman-Pearson Criterion 4
2.2 Problem Formulate 5
2.3 Structure of the Optimal NP Decision Maker 7
2.4 Demspter-Shafer Theory 10
Chapter 3 Proposed Detector Utilizing D-S Theory of Evidence 13
3.1 Specific Decision Problem of Practical Interest: The Gaussian Shift-in-mean 13
3.2 Basic Probability Assignment Estimation 16
3.2.1 BPA Estimation of X Part 16
3.2.2 BPA Estimation of δ Part 20
3.3 The Improved D-S Fusion Algorithm 22
3.3.1 Evaluation of Weighted Factors 23
3.3.2 Credibility Combination Based on D-S Theory 25
3.3.3 Decision Strategy 26
3.3.4 Analysis and Restriction 27
Chapter 4 Simulation Results and Discussions 30
4.1 Comparison of ROC Curves with Different SNR 30
4.2 Comparison of ROC Curves with Different Sample Number 34
4.3 The Relation between the Significance Level and the False Alarm Probability 36
Chapter 5 Conclusion and Future Research 38
REFERENCES 40

[1] S. Marano, V. Matta, and F. Mazzarella, “Refining Decisions After Losing Data: The Unlucky Broker Problem,” IEEE Trans. on Signal Processing, vol.58, no.4, April 2010.
[2] S. Marano, V. Matta, and F. Mazzarella, “The Bayesian Unlucky Broker,” 18th European Signal Processing Conference, Aalborg, Denmark, August 2010.
[3] G. Shafer, A Mathematical Theory of Evidence, Princeton, NJ: Princeton Univ. Press, 1976.
[4] T. Denoeux, “A Neural Network Classifier Based on Dempster-Shafer Theory,” IEEE Trans. on Systems, Man and Cybernetics, vol.30, no.2, pp.131-150, Mar. 2000.
[5] B.V. Dasarathy, “Decision Fusion Strategies in Multi-sensor Environments,” IEEE Trans. on Systems, Man and Cybernetics, vol.21, no.5, pp.1140-1154, Sep/Oct. 1991.
[6] Hongwei Zhu, Basir O., and Karray F., “Data Fusion for Pattern Classification via the Dempster-Shafer Evidence Theory”, IEEE International Conference on Systems, Man and Cybernetics, vol.7, Oct. 2002.
[7] Huadong Wu, Siegel, M., Stiefelhagen, R., and Jie Yang,“Sensor Fusion Using Dempster-Shafer Theory”, IEEE Instrumentation and Measurement Technology Conference, Vol.1 7-12, May 2002.
[8] Qing Yan, and Blum, R.S., “Distributed Signal Detection Under The Neyman-Pearson Criterion,” IEEE Trans. on Information Theory, vol.47, pp. 1368-1377, May 2001.
[9] H. L. Van Trees, Detection, Estimation, and Modulation Theory. Part I. New York: Wiley, 2001.
[10] H. V. Poor, An Introduction to Signal Detection and Estimation. NewYork: Springer-Verlag, 1988.
[11] S. M. Kay, Fundamentals of Statistical Signal Processing, Volume II: Detection Theory. Englewood Cliffs, NJ: Prentice-Hall, 1998.
[12] Karl Sentz and Scott Ferson, “Combination of Evidence in Dempster-Shafer Theory”, Sandia Report SAND2002-0835 Unlimited Release, 13-15, 2002.
[13] P. Smets, “The combination of evidence in the transferable belief model,” IEEE Trans. Pattern Anal. Machine Intell., vol. 12, pp. 447–458, 1990.
[14] S. L. Hegarat-Mascle, D. Richard, and C. Ottle, “Multi-scale data fusion using Demspter-Shafer evidence theory,” Integrated Computer-Aided Engineering, vol. 10, pp. 9–22, 2003.
[15] D.L. Sackett, “Why randomized controlled trials fail but needn't: 2. Failure to employ physiological statistics, or the only formula a clinician-trialist is ever likely to need (or understand!),” CMAJ 165 (9): pp.1226-1237, Oct. 2001.
[16] Z. Wentao, F. Tao, and J. Yan, “Data fusion using improved Dempster Shafer evidence theory for vehicle detection,” in Proc. 4th Int. Conf. on Fuzzy Systems and Knowledge Discovery, vol. 1, Aug. 2007, pp. 487-491.
[17] Hongyu Chen, , and Jian Liu, “Cooperative spectrum sensing based on double threshold detection and Dempster-Shafer theory’’, Communication Technology (ICCT), 2010 12th IEEE International Conference on, pp. 1212-1215, Nov. 2010.