Mobility-aware pilot assignment for UAV-only CF-mMIMO networks

This is the authors' abstract. We don't add key points for this paper.

Unmanned aerial vehicles (UAVs) introduce severe pilot contamination challenges in cell-free massive multiple-input multiple-output (CF-mMIMO) networks due to high mobility and rapidly varying channel conditions. This paper proposes a mobility-aware pilot assignment framework for UAV-only CF-mMIMO systems. The proposed approach combines Kalman-filter-based trajectory prediction, mobility-aware reward design, and interference-aware greedy pilot assignment to proactively adapt pilot allocation based on predicted channel evolution. Simulation results show that the proposed scheme outperforms random and reactive pilot-assignment baselines while achieving SINR performance comparable to the Hungarian-inspired interference-aware benchmark. In particular, the proposed method improves the 5%-likely SINR by 5.91 dB and reduces serving-AP handover events.