Subspace-Enhanced Data-Aided Iterative Channel Estimation for Cell-Free Massive MIMO Systems

Abstract

Cell-free massive multiple-input multiple-output (CF m-MIMO) is a pivotal wireless transmission technology for sixthgeneration (6G) communication systems. However, as the density of user equipment (UE) increases, the sharing of limited orthogonal pilot resources among a large number of UEs leads to severe pilot contamination during channel estimation. Additionally, the strong channel correlation between neighboring UEs exacerbates the difficulty of mitigating this interference. In this paper, a novel subspace-enhanced data-aided (SEDA) iterative channel estimation scheme is proposed for CF m-MIMO systems subject to pilot contamination. By leveraging alternating iterations between channel estimation and signal detection, the pilot contamination compounded by strong inter-UE channel correlation is mitigated through the decision-feedback-equalization principle. In addition, this suppression is further enhanced by projecting the coarse dataaided channel estimate onto the signal subspace. Moreover, by exploiting the access point cooperation, central processing units utilize macro diversity gain for joint signal detection, ensuring rapid algorithm convergence within a few iterations. Simulation results confirm that the proposed SEDA scheme outperforms the state-of-the-art schemes in terms of normalized mean square error of channel estimation, bit error rate, pilot overhead and interference robustness.