Abstract:

Loss of the array aperture and complicated computation have always been serious problems for parameter estimation of the near-field sources. A new array model is proposed which is similar to the NEST array model, using the forth-order cumulation to extend the array aperture. The number of sources estimated by the proposed algorithm is thus much larger than the number of actual arrays. Moreover, we combine the idea of the characteristic equation-based method (CEM) to avoid eigenvalue decomposition and spectrum peak search, greatly reducing the complexity of computation. Finally, based on the angle information of near-field sources, the range information of near-field sources is estimated. Theoretical analysis and simulation results show that on the premise of the accuracy of parameter estimation, the proposed algorithm has lower computational complexity and higher utilization of arrays, which also avoids two-dimensional parameter matching.