Enhancing resilience in cell-free massive MIMO networks via 2D-DOA-based zero-forcing precoding under pilot contamination

Abstract Cell-free massive MIMO networks offer significant advantages in spectral and energy efficiency due to their macro-diversity and distributed architecture. However, the resilience of such systems is challenged by pilot contamination and multi-user interference, particularly in dense deployments where pilot reuse is inevitable. This study proposes a robust and scalable zero-forcing precoding technique based on two-dimensional direction-of-arrival (2D-DOA) estimation to improve network reliability and interference suppression without requiring channel state information (CSI) exchange among access points (APs) or additional pilot overhead. The zero-forcing precoding is aided with DOA information to separate between various users by exploiting the spatial diversity of the correlated channels or allowing spatial multiplexing by antenna weights adaptation based on the characteristics of the channels. By leveraging 2D-UESPRIT and 2D-FDLSM algorithms, the proposed approach mitigates both intra- and inter-cell interference, enhancing system resilience against pilot contamination. A closed-form expression for downlink spectral efficiency is derived, accounting for practical limitations such as imperfect CSI. Simulation results show that the proposed method achieves near-optimal performance—reaching up to 99.1% of the spectral efficiency of a deterministic benchmark based on ideal, interference-free conditions—while significantly outperforming conventional systems under pilot contamination. These findings demonstrate that integrating 2D-DOA-based precoding enhances the robustness and adaptability of cell-free massive MIMO systems, contributing to the development of resilient wireless networks capable of sustaining high performance under real-world constraints.