Source Association Research Articles

Coherency structure exploitation in DOA estimation for non‐circular sources

This study addresses the problem of direction-of-arrival (DOA) estimation for non-circular (NC) sources in the presence of multipath propagation. Recent research reveals that the use of the association of incident signals, i.e. source association (SA), can provide a significant gain in reducing DOA estimation error, especially in the presence of multiple sources. Here, in order to make full use of both the properties of NC sources and the SA information, the multipath model with NC sources is studied, and then a simple combination algorithm is proposed to estimate the SA, based on which, the authors derive an iterative root estimator to refine a given coarse DOA estimation. Meanwhile, the stochastic Cramer-Rao bound taking these two factors into account is derived as well. The proposed method is search-free and maintains the array aperture which is up to twice of the physical array aperture. Numerical simulations have been provided to verify the effectiveness of the proposed method.

Source Association, DOA, and Fading Coefficients Estimation for Multipath Signals

This paper addresses the source association (SA), direction of arrival (DOA), and fading coefficients (FCs) estimation problem in multipath environment. First, we establish a rank reduction property for a multipath signal model with the existence of multiple groups of coherent signals. Subsequently, based on this property, effective algorithms for SA, DOA, and FCs estimation have been developed. The proposed DOA and FCs estimation methods exploit the multipath structure information to achieve improved accuracy. The new DOA estimation methods work well even in the case that the DOAs of the multipath signals associated with different sources are (nearly) overlapped. Meanwhile, the new methods are applicable to arbitrary array geometry while without decreasing the effective array aperture. Then, the stochastic Cramer–Rao bound on DOA and FCs estimation of multipath model (MCRB) exploiting the multipath structure information is derived in closed form. Numerical simulations have been provided to demonstrate the effectiveness of the proposed methods.