A Symmetric Beam Training Method for Near-Field XL-MIMO Systems
With the development of extremely large-scale multiple-input multiple-output systems, there is an increasing demand for highly directional beams in communication systems, making beam training a key technique to ensure efficient communication. However, existing near-field exhaustive beam training schemes require very large training overheads, which is infeasible in practical communication networks. To address this problem, we propose a symmetric beam training scheme for near-field to achieve better rates and reduce the near-field beam training overhead. Specifically, we generate a pair of narrow beams with spatial angle symmetry using a classical polar-domain codebook. Then, simultaneously train the two beams, and the direction and distance of the user are determined by the maximum value of the received power. Simulation results show that the proposed scheme achieves better achievable rate performance than the existing beam training schemes.
Authors
- Jingyi Guan
- Yang Wang (ORCID: https://orcid.org/0000-0001-9438-6060)
- Ke Huang (ORCID: https://orcid.org/0000-0003-1854-7840)
- X. Luo (ORCID: https://orcid.org/0000-0002-4919-303X)
Institutions
- Communication University of China (CN)
Publication Details
- Published
- 2025-04-18
- DOI
- https://doi.org/10.1109/eeice65049.2025.11033820
- Citations
- 1
- Primary Topic
- Advanced MIMO Systems Optimization
- Type
- article
- Field-Weighted Citation Impact
- 0.73