Spatial Spectrum Estimation of Incoherently Distributed Sources Based on Low-rank Matrix Recovery
Résumé
This paper considers the problem of twodimensional(2-D) angle estimation of incoherently distributed(ID) sources in massive MIMO systems. We first show that thematrix representing the 2-D spatial spectrum of ID sources is oflow-rank rather than sparse. Then, by exploiting the low-rankproperty, a rank minimization problem is formulated to estimatethe spatial spectrum which is solved by an efficient iterative reweightedprocedure. Finally, the key parameters of the spatialspectrum are obtained via an off-grid estimator. The performanceanalysis of the proposed method is given, and the Cramer-Raobound (CRB) of the key parameters with unknown angulardistributions is also presented. Contrary to the existing methodswhich need to presume a specific type of angular distribution,the proposed method does not require such presumption, thus issuitable for general angular distribution as long as it is of lowrank.Moreover, the proposed method is more computationallyefficient than the conventional methods that require a multidimensionalsearch. Extensive simulations are provided to verifythe performance of the proposed method.