Performance Analysis of SBL-based Algorithms for DOA Estimation of Smart Antenna Arrays

Smart antenna arrays encompass smart signal processing algorithms suitable for spatial signals detection. To detect and locate the desired signal, various sparse Bayesian learning (SBL) based direction-of-arrival (DOA) estimation algorithms are used. These algorithms require the computation of the inverse of a large matrix for every iteration consequential to high computational complexity. However, this complexity can significantly be reduced by transforming the complex-valued problems into real-valued problems in the uniform linear arrays (ULA). Normally, the accuracy of DOA estimators decreases if the target moves relative to the detector because the motion results in an unknown phase shift of the baseband signal due to the Doppler effect. Therefore, the Doppler frequency must be estimated in addition to the DOA. This paper discusses and analyzes the most recent SBL-based methods challenges and explores novel research directions. Our focus will be on fast SBL-based DOA estimators for fast-moving targets.