Open AccessTraining data selection method for adaptive beamforming
Author(s) -
Dai Baoquan,
Wang Tong,
Bai Tao,
Wu Jianxin,
Bao Zheng
Publication year - 2014
Publication title -
electronics letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.375
H-Index - 146
ISSN - 1350-911X
DOI - 10.1049/el.2014.2024
Subject(s) - adaptive beamformer , beamforming , computer science , snapshot (computer storage) , covariance matrix , training set , covariance , algorithm , machine learning , artificial intelligence , pattern recognition (psychology) , mathematics , statistics , telecommunications , operating system
A method for adaptively selecting training data is proposed to improve the performance of adaptive beamforming. The method first measures the contribution of each snapshot to the covariance matrix required by the beamforming using the sparse iterative covariance‐based estimation technique. Then, those snapshots making larger contributions are selected as the final training samples. The homogeneity of training samples can be improved significantly, and thus results in an evident performance improvement in adaptive beamforming. Simulation results demonstrate the effectiveness of the proposed method.