An information fusion direct position determination method based on Wishart random matrix asymptotic distribution theory
Ren Yanqing, Bin Ba, Zhiyu Lu, Daming Wang · 2017
The traditional multiple-station direct position determination method suffers location accuracy loss and source resolution degradation for the lack of position information fusion of raw data. And an information fusion direct position determination method based on Wishart random matrix asymptotic distribution theory is proposed to overcome the above-mentioned shortcomings. Firstly, the information fusion direct position determination model is established via fusing raw data of each station. Then the new cost function containing eigenspace is constructed with theory of Wishart random matrix asymptotic distribution. Finally, the target location estimation is obtained by two-dimensional geographic grid search. Furthermore, the Cramer-Rao bound of the new model is derived. Compared with the original method, the proposed method performs much better in location accuracy and source resolution by simulations. And it frequently outperforms the information fusion direct position determination method with the cost function only containing noise subspace, under scenarios of low SNR and snapshot deficiency. Its performance has been greatly improved at the expense of lower complexity.