论文信息 - A variable parameter efficient μ-law improved proportionate affine projection algorithm

A variable parameter efficient μ-law improved proportionate affine projection algorithm

The proportionate affine projection algorithm (APA) was developed for echo cancellation. To further improve the convergence rate and tracking ability, in this paper, a variable parameter μ of the μ-law compression is applied to a recently μ-law “memory”-improved proportionate APA (MMIPAPA). In proposed algorithm, the parameter μ is not constant, but time varying and adaptive. Compared with MMIPAPA, simulation results show that the proposed algorithm, termed the adaptive μ-law “memory”-improved proportionate APA (AMMIPAPA), achieves better convergence performance and owns better tracking ability.

Xiaojun Jing | Songlin Sun | Hai Huang | Xiao Xia

[1] Kevin T. Wagner,et al. Gain allocation in proportionate-type NLMS algorithms for fast decay of output error at all times , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.

[2] Donald L. Duttweiler,et al. Proportionate normalized least-mean-squares adaptation in echo cancelers , 2000, IEEE Trans. Speech Audio Process..

[3] Jacob Benesty,et al. An Efficient Proportionate Affine Projection Algorithm for Echo Cancellation , 2010, IEEE Signal Processing Letters.

[4] Jingdong Chen,et al. Acoustic MIMO Signal Processing , 2006 .

[5] Jie Yang,et al. Efficient μ-law improved proportionate affine projection algorithm for echo cancellation , 2011 .

[6] Milos Doroslovacki,et al. Improving convergence of the PNLMS algorithm for sparse impulse response identification , 2005, IEEE Signal Processing Letters.

[7] Kazuhiko Ozeki,et al. An adaptive filtering algorithm using an orthogonal projection to an affine subspace and its properties , 1984 .

[8] Akihiko Sugiyama,et al. A generalized proportionate variable step-size algorithm for fast changing acoustic environments , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.