电子学报2012，Vol.40Issue(6)：1121-1125,5.DOI:10.3969/j.issn.0372-2112.2012.06.009
旋转变换和欠定盲分离的DUET算法
Rotation Transform and DEUT in Underdetermined Blind Source Separation
摘要
Abstract
DUET is a famous algorithm to resolve underdetermined problem using time frequency mask Effect on rotation transform on DUET is discussed and an improved DUET algorithm is proposed in this paper. The algorithm rotates the mixtures and mixing matrix using the rotation matrix, which is formed by any two columns in the mixing matrix, and then do DUET algorithm.Because the results are different in DUET algorithms under different rotations,the algorithm sums the results to improve the distortion.Finally,several speech signal experiments demonstrate its performance and practice.关键词
欠定盲分离/稀疏成分分析/不重叠/时频掩码/l0范数Key words
underdetermined blind source separation/sparse component analysis/disjointness/time frequency mask/l0-norm分类
信息技术与安全科学引用本文复制引用
高峰,肖明,孙功宪,谢胜利..旋转变换和欠定盲分离的DUET算法[J].电子学报,2012,40(6):1121-1125,5.基金项目
国家自然科学基金(No.61003170,No.61174113) （No.61003170,No.61174113）
广东省教育部产学研结合项目(No.2011A090200088) （No.2011A090200088）
广东省自然科学基金重点项目(No.S2011020002735) （No.S2011020002735）
广东茂名市科技计划项目(No.2011B01041) （No.2011B01041）
