本論文提出一個可實現於稀疏陣列的聲源定位和信號擷取的技術，利用直接方法(Direct methods)和基於到達時間差（time difference of arrival, TDOA）的方法比較定位精準度。對於直接方法，在具有稀疏部署的陣列裡，多信號分類（multiple signal classification , MUSIC）演算法比其他直接方法的演算法更不易受空間混疊問題的影響，此外可以利用粒子群最佳化(particles swarm optimizer, PSO)演算法去更有效地定位源位置。對於到達時間差的方法，基於子空間的方法可以比廣義相變互相關（generalized cross correlation with phase transformation, GCC-PHAT）演算法更能精確地估計時間差。當得到時間差的資訊後，我們可以利用受限制的最小平方法（constrained least squares, CLS）做聲源定位，一旦找到聲源位置，就可以使用最小變異無失真響應（minimum variance distortionless response, MVDR）演算法來擷取聲源信號，此外我們還可以應用後濾波器來減少背景噪音，最後進行實驗和客觀的測試來驗證所提出的聲源定位與信號擷取的技術。聲源定位的結果指出直接方法明顯優於其他方法，另外信號提取的部分，在吵雜的環境下音頻質量感知評估（Perceptual evaluation of audio quality , PEAQ）結果指出在所有方法中MVDR過後濾波器最能夠增強信號的品質。
關鍵詞—聲源定位、到達時間、差粒子群最佳化、音訊擷取

This thesis investigates acoustic source localization and separation techniques for sparse arrays. Direct methods and a TDOA-based method are compared in terms of localization accuracy. For the direct localization methods, multiple signal classification (MUSIC) algorithm is less susceptible to spatial aliasing problems than the other direct methods for arrays with sparse deployment. In addition, particles swarm optimizer (PSO) algorithm is applied to efficiently locate source positions. Time delay needs to be estimated prior to the application of the TDOA-based method. A subspace-based method is implemented in this work to estimate the time delays with improved accuracy than the generalized cross correlation with phase transformation (GCC-PHAT) algorithm. After obtaining the TDOA, we can apply the constrained least squares (CLS) algorithm to locate the source. Once the source is located, the minimum variance distortionless response (MVDR) beamformer is utilized to extract source signals, followed by a postfilter to reduce the background noise. Experiments and objective tests are undertaken to validate the direct method and the TDOA-based method for source localization and signal extraction technique. The localization results have demonstrated that the direct method performs better than the TDOA-based method. Perceptual evaluation of audio quality (PEAQ) scores also show that the quality of source signals extracted by using the MVDR beamformer with a minimum mean square error (MMSE) postfilter is significantly enhanced, even in low SNR scenarios.
Index Terms—Source localization, TDOA, PSO, Signal extraction

摘 要 II
ABSTRACT III
誌 謝 IV
LIST OF FIGURES VII
LIST OF TABLES VIII
Chapter 1 INTRODUCTION 1
Chapter 2 ARRAY MODEL 3
Chapter 3 DIRECT METHODS FOR SOURCE LOCALIZATION 5
3.1 MPDR 5
3.2 MUSIC 6
3.3 Modified particles swarm optimizer (MPSO) 7
Chapter 4 TODA-BASED SOURCE LOCALIZATION 8
4.1 TDOA estimation 9
4.1.1 GCC-PHAT 9
4.1.2 Subspace-based method 10
4.2 TDOA measurement model 12
4.3 Constrained Least Squares (CLS) method 14
Chapter 5 SIGNAL EXTRACTION 16
Chapter 6 POSTFILTERING 17
Chapter 7 SIMULATIONS 22
7.1 Delay estimation 22
7.2 Beampatterns of the direct methods 26
7.3 Localization error pattern 29
7.3.1 Localization error for single source 33
7.3.2 Localization error for two sources 34
7.4 Signal extraction 35
Chapter 8 EXPERIMENTS 36
8.1 Source localization 37
8.2 Signal Extraction 38
Chapter 9 CONCLUSIONS 40
REFERENCES 41

[1] R. A. Altes, ‘‘Target position estimation in radar and sonar, and generalized ambiguity analysis for maximum likelihood parameter estimation, ’’ Proceeding of the IEEE, vol. 67, no. 6, pp. 920-930, 1979.
[2] H. Krim, and M. Viberg, ‘‘Two decades of array signal processing research: the parametric approach,” IEEE Signal Processing Magazine, vol. 13, no. 4, pp. 67-94, 1996.
[3] N. Patwari, J. N. Ash, S. Kyperountas, A. O. Hero, R. L. Moses, and N. S. Correal, ‘‘Locating the nodes: cooperative localization in wireless sensor networks,” IEEE Signal Processing Magazine. vol. 22, no.4, pp. 54-69, 2005.
[4] G. Mao, B. Fodan, and B. D. O. Anderson, ‘‘Wireless sensor network localization techniques,” Computer Networks, vol. 51, no.10, pp. 2529-2553, 2007.
[5] M. S. Brandstein, and H. F. Silverman, ‘‘A practical methodology for speech source localization with microphone arrays,” Computer Speech & Language, vol. 11, no.2, pp. 91-126, April 1997.
[6] J. H. DiBiase, H. F. Silverman, and M. S. Brandstein, ‘‘Robust localization in reverberant rooms,” in Microphone Arrays, M.S. Brandstein, and D. Ward (ed.), Springer, Berlin, Heidelberg, 2001.
[7] W. K. Ma, B. N. Vo, S. S. Singh, and A. Baddeley, ‘‘Tracking an unknown time-varying number of speakers using TDOA measurements: a random finite set approach,” IEEE Transactions on Signal Processing, vol. 54, no.9, pp. 3291-3304, September 2006
[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 , Y. T. Chan , and F. K. W. Chan , “ Closed - form formulae for optimum time difference of arrival based localization , ” IEEE Transactions on Signal Processing, vol. 56, no. 6, pp. 2614 – 2620, June 2008.
[11] K. W. Cheung, H. C. So, W.-K. Ma, and Y. T. Chan, “ A constrained least squares approach to mobile positioning: algorithms and optimality, ” EURASIP Journal on Advances in Signal Processing, vol. 2006, Article ID 20858, pp. 1-23, April 2006.
[12] K. W. K. Lui and H. C. So, “A study of two - dimensional sensor placement using time - difference -of - arrival measurements,” Digital Signal Processing, vol. 19, no. 4, pp. 650 – 659, 2009.
[13] H. C. So, “Source localization: Algorithms and analysis,” Handbook of Position Location: Theory, Practice and Advances, Chapter 2, S. A. Zekavat and M. Buehrer, Eds., Wiley-IEEE Press, 2011.
[14] K. W. Cheung, H. C. So, W.-K. Ma and Y. T. Chan, “Least squares algorithms for time-of-arrival based mobile location,” IEEE Transactions on Signal Processing, vol.52, no.4, pp.1121-1128, 2004.
[15] H. C. So and L. Lin, “Linear least squares approach for accurate received signal strength based source localization,” IEEE Transactions on Signal Processing, vol.59, no.8, pp.4035-4040, 2011.
[16] R. Schmidt, "Multiple emitter location and signal parameter estimation,'' IEEE Transactions on Antennas and propagation, vol. 34, pp. 276 280, 1986.
[17] 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
[18] B. V. D. Broeck, A. Bertrand, P. Karsmakers, B. Vanrumste, H. V. Hamme, M. Moonen, “Time-domain generalized cross correlation phase transform sound source localization for small microphone arrays,” IEEE Conference Publications, Amsterdam, Netherlands, pp. 76-80, September 2013.
[19] C. Knapp, G. Carter, "The generalized correlation method for estimation of time delay,'' IEEE Transactions on Acoustics Speech and Signal Processing, vol. 24, no. 4, pp. 320-327, 1976.
[20] J. O. Smith and J. S. Abel, “Closed-form least-squares source location estimation from range-difference measurements,” IEEE Transactions on Acoustics, Speech, and Signal Processing, vol. 35, no.12, pp. 1661-1669, December 1987.
[21] Y. T. Chan and K. C. Ho, “A simple and efficient estimator for hyperbolic location,” IEEE Trans. Signal Processing, vol. 42, no.8, pp. 1905-1915, August 1994.
[22] Y. Huang, J. Benesty, G. W. Elko, and R. M. Mersereau, “Real-time passive source localization: A practical linear-correction least-squares approach,” IEEE Transactions Speech, Audio Processing, vol. 9, no.8, pp. 943-956, November 2001.
[23] P. Loizou, “Speech Enhancement: Theory and Practice,” 2nd. New York, NY, USA: CRC Taylor & Francis, pp. 213-286, 2007.
[24] M. R. Bai, J. G. Ih, and J. Benesty, Acoustic Array Systems: Theory, Implementation, and Application, Wiley-IEEE Press, 1st edition, Singapore, Chaps.3-4, 2013.
[25] J. Kennedy and R. C. Eberhart, “Particle swarm optimization,” IEEE Proceedings of International Conference on Neural Networks, Perth, WA, Australia, Australia, pp. 1942–1948, 1995.
[26] 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.
[27] C. Knapp, G. Carter, "The generalized correlation method for estimation of time delay,'' IEEE Transactions on Acoustics Speech and Signal Processing, vol. 24, no.4, pp. 320-327, 1976.
[28] S. Chen, J. Montgomery, “A Simple Strategy to Maintain Diversity and Reduce Crowding in Particle Swarm Optimization,” AI 2011: Advances in Artificial Intelligence, Perth, Australia, pp. 281-290, December 2011
[29] P. C. Chen, “A non-line-of-sight error mitigation algorithm in location estimation,” IEEE Wireless Communications and Networking Conference, New Orleans, LA, vol. 1, pp. 316-320, August 1999
[30] ITU-T Recommendation P.862, “Perceptual evaluation of speech quality (PESQ): An objective method for end-to-end speech quality assessment of narrow-band telephone networks and speech codecs,” International Telecommunication Union, Geneva, 21pages, 2001.