Entropy-based covariance determinant estimation
Document typeConference lecture
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Rights accessRestricted access - publisher's policy
An information-theoretic approach is described to estimate the determinant of the covariance matrix of a random vector sequence (a common task in a wide range of estimation and detection problems in signal processing for communications). The method is based on a prior entropy-based processing of the data using kernels and offers robustness against small-entropy contamination. The trade-off between optimality, accuracy and robustness is analyzed, along with the impact of the relative kernel bandwidth and data size.
CitationDe Cabrera, F., Riba, J., Vazquez, G. Entropy-based covariance determinant estimation. A: IEEE International Workshop on Signal Processing Advances in Wireless Communications. "The 18th IEEE International Workshop on Signal Processing Advances in Wireless Communications took place during 3-6 July 2017 in Sapporo, Japan". Institute of Electrical and Electronics Engineers (IEEE), 2017, p. 1-5.