A grid of orthogonal beams with zero column coherence can be easily constructed to cover all prospective user equipments (UEs) in the far-field region of a multiple-antenna base station (BS). However, when the BS is equipped with an extremely large aperture array, the Fraunhofer distance is huge, causing the UEs to be located in the radiative near-field region. This calls for designing a grid of beams based on a near-field dictionary. In the previous work, a polar-domain grid design was proposed to maintain control over the column coherence. A limitation of this approach is identified in this paper, and we propose an enhanced methodology for the design of a polar-domain dictionary specifically tailored for the near-field of an extremely large aperture uniform planar array. Through simulation results, it is demonstrated that the proposed dictionary, employing a non-uniform distance sampling approach, achieves lower column coherence than the benchmark and significantly improves the localization of UEs compared to uniform distance sampling.

翻译：在多天线基站远场区域，可通过构建具有零列相干性的正交波束栅格覆盖所有潜在用户设备。然而，当基站配备超大孔径阵列时，夫琅禾费距离显著增大，导致用户设备处于辐射近场区域。这要求基于近场字典设计波束栅格。现有研究提出了极域栅格设计方法以控制列相干性。本文指出该方法的局限性，并提出一种针对超大孔径均匀平面阵列近场场景的增强型极域字典设计方法。仿真结果表明，采用非均匀距离采样策略的所提字典在实现更低列相干性的同时，相比均匀距离采样方案显著提升了用户设备定位性能。