Variable step-size affine projection algorithm based on global speech absence probability for adaptive feedback cancellation

Abstract

A novel approach is proposed for improving adaptive feedback cancellation using a variable step-size affine projection algorithm (VSS-APA) based on global speech absence probability (GSAP). The variable step-size of the proposed VSS-APA is adjusted according to the GSAP of the current frame. The weight vector of the adaptive filter is updated by the probability of the speech absence. The performance measure of acoustic feedback cancellation is evaluated using normalized misalignment. Experimental results demonstrate that the proposed approach has better performance than the normalized least mean square (NLMS) and the constant step-size affine projection algorithms.

Keywords

adaptive feedback cancellation / affine projection / global speech absence probability (GSAP)

References

[1] KimN, ToonW, MadsG C, MarcM, SorenH J. Improved prediction error filters for adaptive feedback cancellation in hearing aids [J]. Signal Processing, 2013, 90(11): 3062-3075

[2] RotaruM, AlbuF, CoandaH. A variable step size modified decorrelated NLMS algorithm for adaptive feedback cancellation in hearing aids [C]. Proc of ISETC. Timisoara, 2012263-266

[3] LiY, ChangL. The application of band-limited NLMS algorithm in hearing aids [C]. Proc of CSIP. Xi’an, 20121446-1449

[4] HuangH C, LeeJ. A new variable step-size NLMS algorithm and its performance analysis [J]. Signal Processing, IEEE Transactions on, 2012, 60(4): 2055-2060

[5] ShinH C, SayedA H, SongW J. Variable step-size NLMS and affine projection algorithms [J]. IEEE Signal Processing Letters, 2004, 11(2): 132-135

[6] LeeC H, ParkP G. Optimal step-size affine projection algorithm [J]. IEEE Signal Processing Letters, 2012, 19(7): 431-434

[7] AlbuF, PaleologuC, CiochinaS. New variable step size affine projection algorithms [C]. Proc of IEEE COMM. Bucharest, 201263-66

[8] OzekiK, UmedaT. An adaptive filtering algorithm using an orthogonal projection to an affine subspace and its properties [J]. Electron Commun Jpn, 1984, 67-A(5): 19-27

[9] FerrerM, GonzalezA, de DiegoM, PineroG. Fast affine projection algorithms for filtered-x multichannel active noise control [J]. IEEE Trans ASLP, 2008, 16(8): 1396-1408

[10] AlbuF, PaleologuC, BenestyJ. A variable step size evolutionary affine projection algorithm [C]. Proc of ICASSP. Prague, 2011429-432

[11] PaleologuC, BenestyJ, CiochinaS. Practical regularization of the affine projection algorithm [C]. Proc of Eusipco. Barcelona, 2011704-708

[12] MayyasK. A variable step-size affine projection algorithm [J]. Digital Signal Processing, 2010, 20(2): 505-510

[13] VegaL R, ReyH, BenestyJ. A robust variable step-size affine projection algorithm [J]. Signal Processing, 2010, 90(9): 2806-2810

[14] GonzalezA, FerrerM, de DiegoM, PineroG. An affine projection algorithm with variable step size and projection order [J]. Digital Signal Processing, 2012, 22(4): 586-592

[15] KimN S, ChangJ H. Spectral enhancement based on global soft decision [J]. IEEE Signal Processing Letters, 2000, 7(5): 108-110

[16] ParkY S, LeeS M. Voice activity detection using global speech absence probability based on teager energy for speech enhancement [J]. IEICE Trans Inf & Syst, 2012, E95-D(10): 2568-2571

[17] HaykinSAdaptive filter theory [R], 2002, Upper Saddle River, NJ, Prentice Hall

[18] EphraimY, MalahD. Speech enhancement using a minimum mean-square error short-time spectral amplitude estimator [J]. IEEE Trans Acoust Speech Signal Process, 1984, 32(6): 1109-1121

[19] SohnJ, KimN S, SungW. A statistical model-based voice activity detection [J]. IEEE Signal Processing Letters, 1999, 6(1): 1-3