Localization of Near-Field Sources Using the Sparse Signal Representation with Symmetric Subarrays

In this paper, based on sparse signal representation, two novel near-field source localization methods are proposed for symmetric subarrays. By utilizing the relationship between the array steering vector of symmetric sensors and separating the two parameters of directions and ranges in propagation time, the near-field source localization problem is converted into a more convenient far-field one. Then the directions and ranges are estimated by sparse-signal-recovery sequentially. The number of sensors of the virtual array using the second method is equavilent to that of the physical array. Thus the number of sources the second method can detect is two times of that while using the first method. The proposed methods show less computation complexity compared to other sparse-signal-recovery methods. Numerical simulation demonstrates that the proposed methods achieve higher resolution ability compared to other methods.