Title :
Multipath TDOA and FDOA estimation using the EM algorithm
Author :
Belanger, Scott P.
Author_Institution :
IBM Federal Systems Co., Owego, NY, USA
Abstract :
The author examines the use of maximum likelihood parameter estimation to determine the time- and frequency difference of arrival (TDOA/FDOA) between direct and multipath signals received at a single sensor. The source signal is assumed to be narrowband, and the additive noise is Gaussian with zero mean. An iterative procedure, based on the estimate-maximize (EM) algorithm, is developed. This procedure permits an unknown complex amplitude and the TDOA/FDOA parameters to be estimated in cases where the source signal and noise spectra are a priori unknowns.<>
Keywords :
iterative methods; maximum likelihood estimation; parameter estimation; random noise; signal processing; time-frequency analysis; EM algorithm; additive noise; complex amplitude; expectation-maximisation algorithm; frequency difference of arrival; iterative procedure; maximum likelihood parameter estimation; multipath signals; time difference of arrival;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1993. ICASSP-93., 1993 IEEE International Conference on
Conference_Location :
Minneapolis, MN, USA
Print_ISBN :
0-7803-7402-9
DOI :
10.1109/ICASSP.1993.319621
