本論文實現隨機排列之麥克風陣列的遠場聲源定位和訊號擷取。此隨機麥克風陣列的位置分佈以模擬退火法（Simulated Annealing, SA）作為最佳化設計。在定位階段，本論文提出區塊粒子群最佳化 (Block Particle Swarm Optimization, BPSO)來有效解決參數估測法 (Parametric method)中多維度的最佳化問題。在模擬使用隨機麥克風陣列的情況下，參數估測法能精確定位在空間中的聲源，有非常好的精準度。
在訊號擷取的階段，首先，我們探討字典學習(Dictionary Learning, DL)中的三個演算法後，以最佳路徑法（Method of Optimal Direction, MOD）評估聲音在空間中的傳遞矩陣（Propagation Matrix）。當聲源的數量小於麥克風數量時則形成過定問題（Overdetermined problem），聲源的振幅便可藉由提可諾夫正規化（Tikhonov Regularization, TIKR）的反算求解。
最後，我們以聲源失真比 (Source-to-Distortion Ratio, SDR) 與聲源干擾比 (Source-to-Interference Ratio, SIR) 作為評比指標來分析兩種方法的擷取成效。

In this thesis, far-field localization and signal extraction using optimized random microphone array is presented. Microphone positions of this random array are optimized with the aid of Simulated Annealing (SA) method. At localization stage, we propose Block Particle Optimization (BPSO) to solve the multi-dimension optimization problem required in parametric approach, the result shows positively in terms of localization accuracy.
In the thesis, several dictionary learning (DL) algorithms are investigated such as the Method of Optimal Direction (MOD) to evaluate the propagation matrix prior for extracting source signals. Source amplitude extraction is achieved by formulating an inverse problem based on the propagation matrix relating the sound pressures received by the microphones and the source amplitudes. The number of sources is selected to be less than the number of microphones to render an over-determined problem, which can be solved by using the Tikhonov regularization (TIKR). To quantify the audio quality of the extracted source signals, Source-to-Distortion Ratio (SDR) and Source-to-Interference Ratio (SIR) are adopted.

摘 要 i
ABSTRACT ii
誌謝 iii
LIST OF TABLES vi
LIST OF FIGURES vii、viii
Chapter 1 INTRODUCTION 1
Chapter 2 FAR-FIELD ARRAY MODEL 5
Chapter 3 PARAMETRIC SOURCE LOCALIZATION 7
3.1 Deterministic maximum likelihood (DML) estimation 8
3.2 Stochastic maximum likelihood (SML) estimation 9
3.3 Parameter estimation in conjunction with Block Particle Swarm Optimization (BPSO) 10
Chapter 4 SOURCE SIGNAL EXTRACTION 13
4.1 Signal Extraction 13
4.2 Propagation matrix optimization with DL algorithms 14
4.2.1 Method of Optimal Direction (MOD)…………………….………..14
4.2.2 K-SVD algorithm………………………………………….………...15
4.2.3 Dictionary Learning using α-Divergence…………………………16
4.3 Procedure of the localization and signal extraction using parametric arrays…………………………………………………………………...…17
Chapter 5 NUMERICAL SIMULATIONS 19
5.1 Parametric source localization……………………………………………19
5.1.1 Localization result…………………………………………………….19
5.1.2 Computational efficiency comparison……………………………….19
5.1.3 Localization in the presence of coherent source…………………….20
5.2 Dictionary Learning algorithms result…………………………………...21
5.3 Source signal extraction result……………………………………………22
Chapter 6 EXPERIMENTAL VALIDATIONS ……………………………….32
6.1 Source localization………………………….………………………….…..32
6.1.1 Source localization using parametric method………………………32
6.1.2 Localization in the presence of coherent source…………………….32
6.2 Source signal extraction……………………………………………….….33
Chapter 7 CONCLUSIONS 41
REFERENCES 42

[1] K. B.Ginn and K. Haddad, “Noise source identification techniques: simple to advanced applications,” Proceedings of the Acoustics, pp. 1781-1786, 2012.
[2] P. Hurley, M. Simeoni, "Flexibeam: Analytic spatial filtering by beamforming," Proc. IEEE ICASSP, pp. 2877-2880, 2016.
[3] Barry D. Van Veen and Kevin M. Buckley, “Beamforming: A Versatile Approach to Spatial Filtering,” IEEE ASSP MAGAZINE, pp. 4-24, April 1988.
[4] H. Krim and M. Viberg, “Two decades of array signal processing research: the Parametric Approach,” IEEE Signal Processing Magazine, vol. 13, July 1996.
[5] S. Seyedin, S. M. Ahadi, "Robust MVDR-based feature extraction for speech recognition," Proceedings of 7th IEEE International Conference on Information Communications and Signal Processing, December 2009.
[6] Y. Kerner and H. Lau, "Two microphone array MVDR beamforming with controlled beamwidth and immunity to gain mismatch," IEEE Int. Workshop Acoust. Signal Enhancement, Aachen, Germany, pp. 1-4, Sep. 4-6, 2012.
[7] G. Mao, B. Fidan, and Brian D.O. Anderson, “Wireless sensor network localization techniques,” Computer Networks, vol. 51, no.10, pp. 2529-2553, 2007.
[8] L. Zhu, and J. Zhu, ‘‘A new model and its performance for TDOA estimation,” IEEE 54th Vehicular Technology Conference, Atlantic City, NJ, USA, USA, pp. 2750-2753, October 2001.
[9] C. Mensing, and S. Plass, ‘‘Positioning algorithms for cellular networks using TDOA,” IEEE International Conference on Acoustics Speech and Signal Processing, Toulouse, France, pp. 513-516, May 2006.
[10] H. C. So and K. W. Chan, “A generalized subspace approach for mobile positioning with time-of-arrival measurements,” IEEE Transactions on Signal Processing, vol.55, no.10, pp.5103-5107, 2007.
[11] E.M. Al-Ardi, R.M. Shubair, and M.E. Al-Mualla, "Computationally efficient high resolution DOA estimation in a multipath environment," IEEE Electron. Lett., Vol. 40, issue 14, pp. 908-910, July 2004.
[12] T.N. Rao and V.S. Rao, "Evaluation of MUSIC algorithm for smart antenna system for mobile communication," Int. Conf. Device, Circuit Syst., Coimbatore, pp. 67-71, March 15-16, 2012.
[13] Ramesh Kawitkar, "Performance of Different types of Array structures based on Multiple signal classification algorithm," IEEE conference on MEMS and Smart Systems, 2009.
[14] N. Ohwada, K. Suyama, "Multiple Sound Sources Tracking Method Based on Subspace Tracking," Proc. IEEE WASPAA 2009, pp. 217-220, October 2009.
[15] A. Sharm and S. Mathur, “Deterministic maximum likelihood direction of arrival estimation using GSA,” IEEE International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT), pp. 415–419, 2016.
[16] Y. Shi et al., “Deterministic maximum likelihood method for direction-of-arrival estimation of strictly noncircular signals,” IEEE Trans. on Acoust., Speech, Signal Proc., 2016
[17] C. En, F. Lorenzelli, R.E. Hudson et al., "Stochastic maximum-likelihood DOA estimation in the presence of unknown nonuniform noise," IEEE Trans. Signal Process., vol. 56, no. 7, pp. 3038-3044, 2008.
[18] P. Stoica et al., “Maximum Likelihood Array Processing for Stochastic Coherent Source,” IEEE Transactions on Signal Processing, vol. 44, no. 1, January 1996.
[19] S. Li, F. Tuteur, "Estimation of underwater source parameters by use of multipath information," The 4th ASSP Workshop on Spectrum Estimation and Modeling, pp. 258-263, 1988.
[20] I. Zislind and M. Wax, “Maximum likelihood localization of multiple sources by alternating projection,” IEEE Trans. On Acoust., Speech, Signal Proc., vol. 36, pp. 1553-1559, 1998
[21] K. C. Sharman, "Maximum likelihood parameter estimation by simulated annealing," Proc. IEEE ICASSP, vol. 88, pp. 2741-2744, 1988.
[22] I. Ziskind, M. Wax, "Maxmimum likelihood localization of diversely polarized sources by simulated annealing," IEEE Trans. Antennas Propag., vol. 38, pp. 1111-1114, 1990.
[23] M. R. Bai, Y. S. Chen and Y, Y, Lo, “A two-stage noise source identification technique based on a far-field random parametric array,” The Journal of the Acoustical Society of America, pp. 2978-2988, 2017.
[24] J. Kennedy and R. Eberhart, “Particle Swarm Optimization,” Proceedings of IEEE International Conference on Neural Networks. IV. pp. 1942–1948, 1995.
[25] R. C. Eberhart and J. Kennedy, “A new optimizer using particle swarm theory,” IEEE Proceedings of the sixth International Symposium on Micro Machine and Human Science, Nagoya, Japan, Japan, pp. 39–43, October 1995.
[26] Z. Wen, D. Goldfarb and K. Scheinberg, “Block Coordinate Descent Methods for Semidefinite Programming,” Handbook on Semidefinite, Conic and Polynomial Optimization pp 533-564. September 2011.
[27] Y. Xu and W. Yin, “Block Coordinate Descent Method for Regularized Multiconvex Optimization with Applications to Nonnegative Tensor Factorization and Completion,” SIAM Journal on Imaging Sciences, vol. 6, no. 3, pp. 1758-1789.
[28] Y. Yang, Z. Li, X. Wang and D. Zhang, “Noise source separation based on the blind source separation,” Chinese Control and Decision Conference, May 2011.
[29] B. Zhao, J. A. Yang, M. Zhang et al., “Research on blind source separation and blind beamforming,” International Conference on Machine Learning and Cybernetics, August 2005.
[30] K. Yao, R. E. Hudson, C. W. Reed, D. Chen, and F. Lorenzelli., “Blind beamforming on a randomly distributed sensor array system,” IEEE J. Sel. Areas Commun., 16, pp. 1555-1567, Oct. 1998.
[31] C. Bao, H. Ji, Y. Quan, Z. Shen, "Dictionary learning for sparse coding: Algorithms and convergence analysis," IEEE Trans. Pattern Anal. Mach. Intell., vol. 38, no. 7, pp. 1356-1369, Jul. 2015.
[32] I. Tosic and P. Frossard, “Dictionary learning,” IEEE Signal Process. Mag., vol. 28, no. 2, pp. 27–38, 2011.
[33] M. Elad, Sparse and Redundant Representations: From Theory to Applications in Signal and Image Processing, New York, NY, USA:Springer, 2010.
[34] M. Aharon, M. Elad, A. Bruckstein, "K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation," IEEE Trans. Signal Process., vol. 54, pp. 4311-4322, 2006.
[35] A. Iqbal and A. Seghouane, "Robust Dictionary Learning Using α-Divergence," IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, United Kingdom, pp. 2972-2976, 2019.
[36] E. Vincent, R. Gribonval and C. Févotte, “Performance Measurement in Blind Audio Source Separation,” IEEE Transaction on Audio, Speech, and Language Processing, vol. 14, NO. 4, July 2006.
[37] E. A. Lehmann. (2010). Fast image-source method: Matlab code [Online]. Available: http://www.eric-lehmann.com/