DoA Estimation of Robust Hybrid Beamforming for Massive MIMO System

This paper presents an in-depth performance analysis of Robust Hybrid Beamforming (RHB) techniques in massive Multiple Input Multiple Output (MIMO) system, emphasizing the integration of adaptive beamforming algorithms to enhance system performance. Focusing on Minimum Variance Distortionless Response (MVDR), Least Mean Squares (LMS), and Linear Constrained Minimum Variance (LCMV) methods, the study evaluates their effectiveness across various scenarios, including different user positions, antenna configurations, and environmental conditions. Through rigorous simulations, the paper demonstrates the superiority of the LMS algorithm in terms of beamwidth, directivity, and interference suppression under standard conditions. However, challenges arise in high-noise environments or when signal and interference directions coincide, prompting the introduction of a Regularized LMS (R-LMS) method to bolster signal direction resilience and maintain performance. Simulation and analytical results shows that the proposed R-LMS based Uniform Rectangular Array (URA) hybrid beamforming model outperforms in normal condition with Side Lobe Level (SLL) of ‐15 dB and Main Lobe Level (MLL) 0 dB. In noisy environment MLL reduced to ‐5 dB but SLL improves to ‐18 dB which provides better noise attenuation. The findings advocate for the potential of RHB in advancing beamforming technologies, promising significant improvements in wireless communication systems' efficiency and reliability.