A Fast Direct Position Determination for Multiple Sources Based on Radial Basis Function Neural Network

Compared with the conventional two-step Iocalization mode, direct position determination (DPD) algorithm avoids the measurement-source association problem in multiple sources scenario, and has the advantages of higher Iocalization accuracy and stronger resolution capability. However, the existing DPD algorithms, e.g. maximum likelihood (ML)-based DPD algorithm, are unsuitable for real-time applications due to high computational complexity. In this paper, a fast DPD method using radial basis function (RBF) neural network (NN) has been proposed. To reduce the dimension of the input space, an effective pre-processing scheme is present. A reliable training process improves the generalization performance of NN. Simulation results show the feasibility of the proposed algorithm and demonstrate that the proposed method is more computationally efficient than the existing ML-based DPD algorithm.