Array Self-Position Determination Based on Orthogonal Grid Matching Under Multipath Environments

Abstract

Array self-position determination methods based on multiple emitter data can avoid significant deviations of vehicle satellite navigation in harsh environments. However, existing array self-position determination methods show decrease in performance under multipath environments. To deal with this problem, we propose an array self-position determination method based on orthogonal grid matching with the spatial differencing method. Specifically, the direction of arrival (DOA) of direct path and multipath signals are respectively estimated by array spatial differencing method. The matching accuracy is enhanced by utilizing the prior information of direct path signal. After calculating correlation coefficients of different sources, estimated angles with high correlation are then classified into the same set. Then, the noise subspace of each angle set is reconstructed and the position is estimated by grid matching with the orthogonal property between the noise subspaces and the characteristic steering vectors. The matching results of redundant angle sets are removed as non-matching items, thus averting positioning deviations. The simulation results demonstrate that the computational complexity of the proposed method is comparable to that of the signal subspace fitting (SSF). Moreover, in terms of positioning precision, the proposed method outperforms multiple signal classification with enhanced spatial smoothing (ESSMUSIC), initial signal fitting (ISF), and SSF.