Title :
A data-based enumeration technique for fully correlated signals
Author :
Krim, Hamid ; Cozzens, John H.
Author_Institution :
Lab. for Inf. & Decision Syst., MIT, Cambridge, MA, USA
fDate :
7/1/1994 12:00:00 AM
Abstract :
Presents a novel method for estimating the number of signals impinging on a uniform linear array using observed sensor data. Unlike other algorithms that apply Rissanen´s minimum description length (MDL) principle to the observed data for source enumeration, this method applies it to the prediction errors of a linear model that has been fitted to an appropriate data matrix. It is a 1D method that achieves improved performance even for fully correlated signals over contemporary approaches, particularly with short data records and closely spaced signals. Asymptotic consistency is shown and substantiating simulation examples are included
Keywords :
array signal processing; error analysis; filtering and prediction theory; linear systems; parameter estimation; signal detection; 1D method; data matrix; data-based enumeration technique; fully correlated signals; linear model; number of signals; performance; prediction errors; sensor data; simulation examples; source enumeration; uniform linear array; Covariance matrix; Direction of arrival estimation; Eigenvalues and eigenfunctions; Multidimensional systems; Predictive models; Sensor arrays; Signal processing; Smoothing methods; Testing; Working environment noise;
Journal_Title :
Signal Processing, IEEE Transactions on
