A multichannel cross-spectrum density estimator based on canonical correlation analysis
Document Type
Conference Proceeding
Publication Date
10-12-2012
Abstract
In this work, the problem of multichannel cross-spectrum density (MCSD) estimation is studied. Based on a multichannel data model, the classic periodogram based MCSD estimator and the minimum variance (MV) based MCSD estimator are tested for cross-spectrum density estimation. Our major contribution in this work is a canonical correlation analysis (CCA) based MCSD estimator, relying on the inherent maximization arguments of CCA. It is demonstrated through a model based multichannel simulation example that the newly proposed CCA-MCSD estimator is of high frequency resolution as well as good sidelobe suppression properties, and can serve as an excellent candidate for MCSD estimation. © 2012 IEEE.
Identifier
84867192913 (Scopus)
ISBN
[9781467310710]
Publication Title
Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop
External Full Text Location
https://doi.org/10.1109/SAM.2012.6250558
e-ISSN
2151870X
First Page
533
Last Page
536
Recommended Citation
Wang, Xiaoli and Ge, Hongya, "A multichannel cross-spectrum density estimator based on canonical correlation analysis" (2012). Faculty Publications. 18069.
https://digitalcommons.njit.edu/fac_pubs/18069
