本論文提出一種基於機器學習對最小相移鍵控調變脈衝串訊息的訊號起始點檢測方法。訊號端點檢測是訊號處理中一個十分重要的技術，主要可以分為特徵提取和閾值設置兩個步驟，通過接收訊號提取出的特徵將訊號從雜訊環境中區分為訊號段與非訊號段，並且使用閾值設置來找出訊號段的起始點和終止點。但在實際應用時常因為環境的不同，造成閾值不易設置，而且不同的訊號也會使設置的條件難以統一。
本論文結合特徵提取與機器學習分類器對最小相移鍵控調變脈衝串的訊息進行訊號起始點檢測。將接收訊息提取出特徵，並且透過預處理技術和特徵選擇方法選擇出評分最好的特徵子集，再經由模型選擇找出最適合訊息的分類器演算法。為此我們可以知道對於最小相移鍵控調變脈衝串訊息，特徵與分類器演算法的最佳組合，以求得最佳化的分類效能。最後透過分類訊息的訊號段與非訊號段，利用演算法找出訊息的起始點。
本論文利用模擬軟體MATLAB模擬最小相移鍵控調變脈衝串的訊息，並且使用Python編程語言以及機器學習軟件庫Scikit-learn中的分類器演算法，進行本論文方法的功能驗證與效能分析，最後探討訊息中含有不同訊雜比對於訊號起始點檢測的影響。模擬結果顯示最小相移鍵控調變脈衝串訊息在不同訊雜比的環境下，使用最佳的特徵與分類器演算法的組合，對於訊號起始點檢測有著較低的估測誤差。

This paper proposes a method for detecting the signal starting point of an MSK modulated pulse train based on machine learning. Signal endpoint detection is a very important technology in signal processing. It can be divided into two steps: feature extraction and threshold setting. The features extracted from the received signal separate the signal from the noise environment into signal and non-signal segments. And use the threshold setting to find the start point and end point of the signal segment. However, in practical applications, it is often difficult to set the threshold due to different environments, and different signals also make it difficult to unify the setting conditions.
This paper combines feature extraction and machine learning classifier to detect the signal starting point of an MSK modulated pulse train. The features of the received message are extracted, and the subset of features with the best score is selected through preprocessing technology and feature selection method, and then the most suitable classifier algorithm for the message is found through model selection. For this reason, we can know the best combination of features and classifier algorithm for the minimum phase shift keying modulation burst message to obtain the best classification performance. Finally, by classifying the signal segment and non-signal segment of the message, an algorithm is used to find the starting point of the message.
This paper uses the simulation software MATLAB to simulate the message of the minimum phase shift keying modulation pulse train, and uses the Python programming language and the classifier algorithm in the machine learning software library Scikit-learn to perform the function verification and performance analysis of the method in this paper. Finally, the influence of different signal-to-noise ratios in the message on the detection of the signal starting point is discussed. The simulation results show that the minimum phase shift keying modulated burst message uses the best combination of features and classifier algorithms in environments with different signal-to-noise ratios, and has a lower estimation error for the signal onset detection.

[1] L. Rabiner and B.-H. Juang, Fundamentals of speech recognition. 1993.
[2] M. H. Savoji, "A robust algorithm for accurate endpointing of speech signals," Speech communication, vol. 8, no. 1, pp. 45-60, 1989.
[3] R. E. Ziemer and W. H. Tranter, Principles of Communications: Systems, Modulation and Noise, 7th ed. 2014.
[4] S. Haykin and M. Moher, Communication Systems, 5th ed. 2017.
[5] C. H. Kao and N. P. School, Performance Analysis of a JTIDS/Link-16-type Waveform Transmitted Over Slow, Flat Nakagami Fading Channels in the Presence of Narrowband Interference. Naval postgraduate school monterey ca, 2008.
[6] N. Grumman, "Understanding voice and data link networking," Accessed on May, vol. 9, p. 2014, 2013.
[7] 梅文华 and 蔡善法, JTIDS/Link16 数据链. 北京: 国防工业出版社, 2007.
[8] L. R. Rabiner and M. R. Sambur, "An algorithm for determining the endpoints of isolated utterances," Bell System Technical Journal, vol. 54, no. 2, pp. 297-315, 1975.
[9] J. Haigh and J. Mason, "Robust voice activity detection using cepstral features," in Proceedings of TENCon'93. IEEE Region 10 International Conference on Computers, Communications and Automation, 1993, vol. 3, pp. 321-324.
[10] J.-l. Shen, J.-w. Hung, and L.-s. Lee, "Robust entropy-based endpoint detection for speech recognition in noisy environments," in Fifth international conference on spoken language processing, 1998.
[11] C. Jia and B. Xu, "An improved entropy-based endpoint detection algorithm," in International Symposium on Chinese Spoken Language Processing, 2002.
[12] H. Liang-Sheng and Y. Chung-Ho, "A novel approach to robust speech endpoint detection in car environments," in 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100), 2000, vol. 3, pp. 1751-1754
[13] B. P. Bogert, "The quefrency alanysis of time series for echoes; Cepstrum, pseudo-autocovariance, cross-cepstrum and saphe cracking," Time series analysis, pp. 209-243, 1963.
[14] C. E. Shannon, "A mathematical theory of communication," Bell system technical journal, vol. 27, no. 3, pp. 379-423, 1948.
[15] L. Rabiner and M. Sambur, "Voiced-unvoiced-silence detection using the Itakura LPC distance measure," in ICASSP'77. IEEE International Conference on Acoustics, Speech, and Signal Processing, 1977, vol. 2, pp. 323-326.
[16] J.-C. Junqua, B. Mak, and B. Reaves, "A robust algorithm for word boundary detection in the presence of noise," IEEE Transactions on speech and audio processing, vol. 2, no. 3, pp. 406-412, 1994.
[17] S. Raschka and V. Mirjalili, Python Machine Learning: Machine Learning and Deep Learning with Python, scikit-learn, and TensorFlow 2. Packt Publishing Ltd, 2019.
[18] S. Ioffe and C. Szegedy, "Batch normalization: Accelerating deep network training by reducing internal covariate shift," 2015.
[19] R. Bellman, "Dynamic programming," Science, vol. 153, no. 3731, pp. 34-37, 1966.
[20] R. E. Bellman, Adaptive control processes: a guided tour. Princeton university press, 2015.
[21] G. Hughes, "On the mean accuracy of statistical pattern recognizers," IEEE transactions on information theory, vol. 14, no. 1, pp. 55-63, 1968.
[22] I. Guyon and A. Elisseeff, "An introduction to variable and feature selection," Journal of machine learning research, vol. 3, pp. 1157-1182, 2003.
[23] A. Bommert, X. Sun, B. Bischl, J. Rahnenführer, and M. Lang, "Benchmark for filter methods for feature selection in high-dimensional classification data," Computational Statistics & Data Analysis, vol. 143, 2020.
[24] V. Spruyt, "The Curse of Dimensionality in classification," Computer vision for dummies, vol. 21, no. 3, pp. 35-40, 2014.
[25] C. Dai, L. Luo, H. Peng, and Q. Sun, "A Method Based on Support Vector Machine for Voice Activity Detection on Isolated Words," in 2018 13th International Conference on Computer Science & Education (ICCSE), 2018, pp. 1-4.
[26] J. Dey, M. S. B. Hossain, and M. A. Haque, "An Ensemble SVM-based Approach for Voice Activity Detection," in 2018 10th International Conference on Electrical and Computer Engineering (ICECE), 2018, pp. 297-300.
[27] A. Géron, Hands-On Machine Learning with Scikit-Learn & TensorFlow. 2017.
[28] C. J. Burges, "A tutorial on support vector machines for pattern recognition," Data mining and knowledge discovery, vol. 2, no. 2, pp. 121-167, 1998.
[29] V. Vapnik, The nature of statistical learning theory. Springer science & business media, 2013.
[30] C. Cortes and V. Vapnik, "Support-vector networks," Machine learning, vol. 20, no. 3, pp. 273-297, 1995.
[31] B. E. Boser, I. M. Guyon, and V. N. Vapnik, "A training algorithm for optimal margin classifiers," in Proceedings of the fifth annual workshop on Computational learning theory, 1992, pp. 144-152.
[32] L. Rokach and O. Maimon, "Top-down induction of decision trees classifiers-a survey," IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), vol. 35, no. 4, pp. 476-487, 2005.
[33] J. Bergstra and Y. Bengio, "Random search for hyper-parameter optimization," Journal of machine learning research, vol. 13, no. Feb, pp. 281-305, 2012.
[34] E. Brochu, V. M. Cora, and N. De Freitas, "A tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning," 2010.
[35] J. Snoek, H. Larochelle, and R. P. Adams, "Practical bayesian optimization of machine learning algorithms," in Advances in neural information processing systems, 2012, pp. 2951-2959.
[36] C. J. Van Rijsbergen, Information retrieval. London; Boston: Butterworths, 1979.
[37] D. E. Knuth, "Seminumerical algorithm," The art of computer programming, 1981.
[38] R. Kohavi, "A study of cross-validation and bootstrap for accuracy estimation and model selection," 1995, vol. 14, no. 2, pp. 1137-1145: Montreal, Canada.