Title :
Bias analysis for ESPRIT-type estimation algorithms
Author :
Li, Fu ; Lu, Yang
Author_Institution :
Dept. of Electr. Eng., Portland State Univ., OR, USA
fDate :
3/1/1994 12:00:00 AM
Abstract :
A statistical analysis using perturbation expansion is applied to the DOA estimation bias for ESPRIT, state-space realization (TAM), and matrix-pencil algorithms, assuming that only a finite amount of array data is available. The authors have obtained an analytical expression for the DOA estimation bias in a simple and self-contained fashion. The tractable formula provides insight to the algorithms. Simulation results verify the analytically predicted performance
Keywords :
array signal processing; matrix algebra; parameter estimation; perturbation techniques; state-space methods; statistical analysis; DOA estimation; ESPRIT-type estimation algorithms; bias analysis; finite array data; matrix-pencil algorithms; performance; perturbation expansion; simulation results; state-space realization; statistical analysis; Algorithm design and analysis; Direction of arrival estimation; Error analysis; Estimation error; Multiple signal classification; Narrowband; Performance analysis; Sensor arrays; State estimation; Statistical analysis;
Journal_Title :
Antennas and Propagation, IEEE Transactions on
