Cm. Cho et Pm. Djuric, DETECTION AND ESTIMATION OF DOAS OF SIGNALS VIA BAYESIAN PREDICTIVE DENSITIES, IEEE transactions on signal processing, 42(11), 1994, pp. 3051-3060
A new criterion based on Bayesian predictive densities and subspace de
composition is proposed for simultaneous detection of signals impingin
g on a sensor array and estimation of their direction-of-arrivals (DOA
's). The solution is applicable for both coherent and noncoherent sign
als and an arbitrary array geometry. The proposed detection criterion
is strongly consistent and outperforms the MDL and AIC criteria, parti
cularly for a small number of sensors and/or snapshots, and/or low SNR
, without increased computational complexity. When the prior of the di
rection-of-arrival is a uniform distribution, the Bayesian estimator f
or the directional parameters coincides with the unconditional maximum
likelihood estimator. Simulation results that demonstrate the perform
ance of the proposed solution are included.