Blind source separation of indoor mobile voice sources

Blind source separation of indoor mobile voice sources

Chunli Wang Quanyu Wang  Yuping Cao 

College of Electronics and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China

Corresponding Author Email: 
wcl@mail.lzjtu.cn
Page: 
179-183
|
DOI: 
10.18280/mmep.040407
Received: 
| |
Accepted: 
| | Citation

OPEN ACCESS

Abstract: 

The mobile voice sources move freely in the indoor range of several to tens of square meters at a speed of lower than 10m/s. The reflection signal and the subsequent original sound signal are superposed in each space to generate reverberation and cause serious interference to the original sound signal. Through the comparison of three classical algorithms of blind source separation, the online algorithm can constantly update the separation system in real time according to the different positions of voice source signals, but it has no advantages in such performances as operation speed and convergence speed. The batch algorithm is fast but delayed, while the blind source separation algorithm based on independent component analysis of the frequency domain has less computation and fast convergence. The improved separation matrix algorithm is used to verify the effectiveness of the algorithm.

Keywords: 

Mobile Voice Sources, Reverberation, Blind Source Separation, Natural Gradient, Independent Component Analysis

1. Introduction
2. Reverberation
3. Blind Source Separation Algorithm for Mobile Speech Signals Based on Frequency Domain ICA
4. Optimization of ICA Blind Source Separation Algorithm
5. Simulation Experiment and Analysis of Results
6. Conclusions
Acknowledgements
  References

[1] Liao T. (2012). Researches of the blind source separation algorithm based on an independent component analysis, Master dissertation of Hunan Normal University, pp. 15-21, DOI: 10.7666/d.y2149207

[2] Zhou G.X., Yang Z.Y., Xie S.L. (2011). Mixing matrix estimation from sparse mixtures with unknown number of sources, IEEE Transactions on Neural Networks, Vol. 22, No. 2, pp. 211-221. DOI: 10.1109/TNN.2010.2091427

[3] Cao Y.P. (2012). Speech enhancement algorithm based on the signal subspace, Electronic Test, Vol. 6, pp. 54-57.

[4] Zhou J. (2016). Research of underdetermined source estimation and blind extraction method for mechanical fault signals, Doctoral dissertation of Kunming University, pp. 38-45.

[5] Yang P. (2008). Blind source separation study on the aero engine aliasing vibration signal, Master dissertation of Nanjing University of Aeronautics and Astronautics, pp. 13-15. DOI: 10.7666/d.d053213

[6] Guo W., Yu F.Q. (2015). Improved speech music signal separation based on negative entropy maximization, Computer Engineering and Application, Vol. 51, No. 4, pp. 209-212,

[7] Lu J.T., Cheng W. (2015). Adaptive blind source separation algorithm in a changed or equivariant step, Journal of Xi 'an Jiaotong University, Vol. 49, No. 12, pp. 83-89. DOI: 10.7652/xjtuxb201512014

[8] Zhang Y.Y., Xin J.H., Liu G.B. (2016). Applications of combined with cumulant slice joint diagonalization of blind source separation, Journal of Huazhong University of Science and Technology (Natural Science), Vol. 44, No. 7, pp. 86-90. DOI: 10.13245/j.hust.160717

[9] Dermoune A., Wei T. (2013). Fast ICA algorithm: five criteria for the optimal choice of the nonlinearity function, IEEE Transactions on Signal Processing, Vol. 61, No. 8, pp. 2078-2087. DOI: 10.1109/TSP.2013.2243440

[10] Chen G.Q. (2015). Blind source separation algorithm based on kurtosis cumulant proportional differential control vector, Journal of Electronics, Vol. 12, No. 5, pp. 929-934. DOI: 10.3969/j.issn.0372-2112.2015.05.015

[11] Yang J.M., Qi H.Y. (2015). Improved nonlinear blind source separation algorithm based on the minimization of mutual information, Electric Measurement and Instrument, Vol. 52, No. 9, pp. 66-69. DOI: 10.3969/j.issn.1001-1390.2015.09.013