Maximum Likelihood Based Channel Delay Estimation with Limited Number of Samples Using Diagonal Matrix Approximation
Document Type
Article
Publication Date
12-1-2003
Abstract
To achieve fast and computationally simplified channel estimation, we propose a maximum likelihood estimation (MLE) based "diagonal matrix approximation" (DMA) scheme which provides an effective solution to overcome the conflict between estimation accuracy, acquisition time, and complexity. Together with over sampling, the proposed scheme exhibits a performance approaching the Cramér-Rao Lower Bound (CRLB), while keeping computational complexity linear to the signal dimensions. The DMA scheme can still perform well even when the sample estimate of the covariance matrix, due to insufficient number of observation samples, is singular. Therefore, the proposed scheme is a viable solution for fast channel estimation.
Identifier
0942278603 (Scopus)
Publication Title
Wireless Personal Communications
External Full Text Location
https://doi.org/10.1023/B:WIRE.0000010146.60266.4e
ISSN
09296212
First Page
183
Last Page
194
Issue
3
Volume
27
Grant
9903381
Fund Ref
National Science Foundation
Recommended Citation
Tang, Xin and Bar-Ness, Yeheskel, "Maximum Likelihood Based Channel Delay Estimation with Limited Number of Samples Using Diagonal Matrix Approximation" (2003). Faculty Publications. 13794.
https://digitalcommons.njit.edu/fac_pubs/13794
