Title :
Using the Pearson correlation coefficient to develop an optimally weighted cross relation based blind SIMO identification algorithm
Author :
Huang, Yiteng ; Benesty, Jacob ; Chen, Jingdong
Author_Institution :
WeVoice, Inc., Bridgewater, NJ
Abstract :
Blind SIMO identification is challenging when additive noise is strong and for ill-conditioned/acoustic SIMO systems. A weighted cross relation (CR) algorithm presumably can be robust to noise but there lacks a practical way to define the weights. In this paper, the Pearson correlation coefficient (PCC) is used to develop an optimally weighted CR algorithm, which is validated by simulations.
Keywords :
acoustic signal processing; blind source separation; correlation methods; noise; Pearson correlation coefficient; acoustic SIMO system; additive noise; blind SIMO identification algorithm; ill-conditioned system; weighted cross relation algorithm; Acoustic noise; Additive noise; Chromium; Finite impulse response filter; Jacobian matrices; Microphones; Noise robustness; Signal processing; Speech; Statistics; Pearson correlation coefficient; Weighted cross relations; acoustic SIMO system; blind identification;
Conference_Titel :
Acoustics, Speech and Signal Processing, 2009. ICASSP 2009. IEEE International Conference on
Conference_Location :
Taipei
Print_ISBN :
978-1-4244-2353-8
Electronic_ISBN :
1520-6149
DOI :
10.1109/ICASSP.2009.4960293
