School of Mathematics and Physics, China University of Geosciences, Wuhan 430074, China

---

Abstract—The sparsity of signals in their transformed domain is widely used for under-determined blind source separation. The most challenging task of under-determined BSS is to estimate the mixing matrix. In this paper, a new cost function is proposed to detect the sparsest sub-band. Samples in the sub-band can be used to estimate the mixing matrix. Finally some numerical experiments are performed to evaluate the effectiveness of the proposed algorithm.

Index Terms—Under-determined blind source separation, sparse component separation, wavelet packet transform, kmeans clustering.

[PDF]

Cite: Yujie Zhang, Huiming Peng, Hongwei Li, "Wavelet Packet Transform-Based Algorithm for Mixing Matrix Estimation," Journal of Computers vol. 7, no. 11, pp. 2605-2611, 2012.