Accurate Iterative Algorithm for Detection and the Signal AoA Estimation in Low SNR Region

The GLRT (Generalized Likelihood Ratio Test) method is applied for joint detection and the angle-of-arrive (AoA) estimation of an unknown signal in multi-element antenna arrays. The direct method of the maximum likelihood (ML) estimation of the power and the AoA of a signal with a plane wavefront and unknown temporal structure is considered. An efficient iterative procedure is proposed for implementation of this approach. For the useful signal detection the Neumann-Pearson criterion is applied at the final iteration step and the threshold value for the decision statistics is calculated by simulations. The performance characteristics of the proposed joint detection-estimation iterative algorithm are compared with three-step algorithm, which based on the sample correlation matrix eigenvectors and eigenvalues analysis. It was found that in high SNR region the accuracies of the proposed iterative algorithm and the three-step algorithm practically reach the low CRBs, but in low SNR region and the short sample sizes the iterative algorithm at 35% - 50% over performs the accuracy of the three-step algorithm.