Tensor-based methods for blind spatial signature estimation under arbitrary and unknown source covariance structure

详细信息    查看全文

• 作者：Paulo R.B. Gomes^a ; ^{paulo@gtel.ufc.br}Author Vitae ; André ; L.F. de Almeida^a ; ^{andre@gtel.ufc.br}Author Vitae ; Joã ; o Paulo C.L. da Costa^b ; ^{joaopaulo.dacosta@ene.unb.br}Author Vitae ; Giovanni Del Galdo^b ; ^{giovanni.delgaldo@iis.fraunhofer.de}Author Vitae
• 关键词：Array processing ; Spatial signature estimation ; Tensor decomposition ; Alternating least squares
• 刊名：Digital Signal Processing
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：62
• 期：Complete
• 页码：197-210
• 全文大小：1144 K
• 卷排序：62

文摘

Higher-order tensor methods for blind spatial signature estimation are proposed. Algorithms able to operate when sources covariance structure is unknown and arbitrary. More efficient denoising based on expanded form of spatial smoothing. Identifiability, uniqueness and computational complexity issues are analyzed.