Translator Disclaimer
Paper
23 July 2003 Design and analysis of supervised and decision-directed estimators of the MMSE/LCMV filter in data-limited environments
Author Affiliations +
Abstract
In this paper we quantify theoretically the effect of the desired-signal power level on the mean square filter estimation error and the normalized output signal-to-interference-plus-noise-ratio (SINR) of sample matrix inversion (SMI)-type estimates of the minimum mean-square-error (MMSE) and the linearly constrained minimum variance (LCMV) filters. We prove that in finite data support situations filters that utilize a sample average estimate of the desired-signal-absent input correlation matrix exhibit superior normalized filter output SINR and mean square filter estimation error when compared to filters that utilize a sample average estimate of the desired-signal-present input correlation matrix. Finally, we investigate pilot-assisted and decision-directed adaptive filter implementations that exhibit near desired-signal-absent SMI-filtering performance while they are trained using desired-signal-present data/observations.
© (2003) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jeffrey M. Farrell, Ioannis N. Psaromiligkos, and Stella N. Batalama "Design and analysis of supervised and decision-directed estimators of the MMSE/LCMV filter in data-limited environments", Proc. SPIE 5100, Digital Wireless Communications V, (23 July 2003); https://doi.org/10.1117/12.487947
PROCEEDINGS
11 PAGES


SHARE
Advertisement
Advertisement
RELATED CONTENT


Back to Top