Near-Field Beamforming Over Non-Stationary Channels for Extremely Large-Scale Arrays

Extremely large-scale arrays and high frequencies are vital technologies for enhancing the capacity in wireless communications. The employment of these technologies makes the near-field propagation become dominant, where the achievable rate is reduced due to the misfocus effect caused by the beam mismatch. To mitigate it, beamforming schemes assuming spherical wavefront propagation must be investigated. However, the integration of angle and distance information in the beam domain, alongside the impacts of molecular absorption and spherical wave reflection coefficients, has been scarcely explored. We derive the expression of the molecular absorption at high frequencies and formulate the reflection coefficient expression under the spherical wave assumption, and embed them into the design of the beamforming. We then analyze these characteristics and propose a near-field beamforming scheme utilizing the Gauss-Legendre quadrature and design a minimum difference mapping strategy for phase mapping. Based on this scheme, we propose two beamforming algorithms for line-of-sight and non-line-of-sight scenarios, respectively. Numerical results demonstrate that the proposed algorithms are robust against the beam misfocus effect and achieve a higher achievable transmission rate.