View PDF HTML (experimental)

Abstract:In this paper, we propose to leverage rotatable antennas (RAs) for improving the communication performance in mixed near-field and far-field communication systems by exploiting a new spatial degree-of-freedom (DoF) offered by antenna rotation to mitigate complex near-field interference and mixed-field interference. Specifically, we investigate a modular RA-enabled mixed-field downlink communication system, where a base station (BS) consisting of multiple RA subarrays communicates with multiple near-field users in the presence of several legacy far-field users. We formulate an optimization problem to maximize the sum-rate of the near-field users by jointly optimizing the power allocation and rotation angles of all subarrays at the BS. To gain useful insights into the effect of RAs on mixed-field communications, we first analyze a special case where all subarrays share the same rotation angle and obtain closed-form expressions for the rotation-aware normalized near-field interference and the rotation-aware normalized mixed-field interference using the Fresnel integrals. We then analytically reveal that array rotation effectively suppresses both interference types, thereby significantly enhancing mixed-field communication performance. For the general case involving subarray-wise rotation, we propose an efficient double-layer algorithm to obtain a high-quality solution, where the inner layer optimizes power allocation using the successive convex approximation (SCA) technique, while the outer layer determines the rotation angles of all subarrays via particle swarm optimization (PSO). Finally, numerical results highlight the significant performance gains achieved by RAs over conventional fixed-antenna systems and demonstrate the effectiveness of our developed joint design compared to benchmark schemes.

Submission history

From: Yunpu Zhang [view email]
[v1] Fri, 5 Sep 2025 07:29:50 UTC (256 KB)
[v2] Wed, 17 Sep 2025 03:45:02 UTC (256 KB)
[v3] Wed, 19 Nov 2025 15:56:13 UTC (734 KB)
[v4] Tue, 25 Nov 2025 13:17:50 UTC (734 KB)
[v5] Sun, 6 Sep 2026 04:51:38 UTC (240 KB)