Hybrid Beamforming Optimization for DOA Estimation Based on the CRB Analysis

TL;DR

This paper proposes a manifold optimization algorithm to design hybrid beamforming matrices that minimize the CRB for DOA estimation in mmWave MIMO systems, significantly improving estimation accuracy.

Contribution

It introduces a CRB-based HBF optimization framework and an efficient MO algorithm to enhance DOA estimation performance in hybrid beamforming systems.

Findings

CRB-MO algorithm outperforms random HBF methods

Significant reduction in CRB achieved

Provides practical insights for beam training design

Abstract

Direction-of-arrival (DOA) estimation is one of the most demanding tasks for the millimeter wave (mmWave) communication of massive multiple-input multiple-output (MIMO) systems with the hybrid beamforming (HBF) architecture. In this paper, we focus on the optimization of the HBF matrix for receiving pilots to enhance the DOA estimation performance. Motivated by the fact that many existing DOA estimation algorithms can achieve the Cram\'{e}r-Rao bound (CRB), we formulate the HBF optimization problem aiming at minimizing the CRB with the prior knowledge of the rough DOA range. Then, to tackle the problem with intractable non-convex constraint introduced by the analog beamformers, we propose an efficient manifold optimization (MO) based algorithm. Simulation results demonstrate the significant improvement of the proposed CRB-MO algorithm over the conventional random HBF algorithm, and provide insights for the HBF design in the beam training stage for practical applications.