.png){kind=icon}
.png){kind=icon}
Return
DANM-ADMM for RIS-Aided Gridless 2-D DOA Estimation in Vehicular NLoS Scenarios
.png){kind=icon}
PRE
AI
Z
Z
T
P
X
DOI：10.1109/lcomm.2026.3719280.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Conventional direction-of-arrival (DOA) estimation methods often depend on an available line-of-sight (LoS) link. In non-line-of-sight (NLoS) scenarios, uncontrolled signal reflections can make target angular information difficult to recover, leading to a clear performance loss. To address this problem, this letter studies a reconfigurable intelligent surface (RIS)-based gridless DOA estimation system, where the RIS forms a virtual LoS link between the sensing node and the targets. We formulate DOA estimation through decoupled atomic norm minimization (DANM), using the spatial-domain sparsity of targets. Since conventional semidefinite programming (SDP)-based solvers for the DANM problem are computationally costly and difficult to implement efficiently, we further propose DANM-ADMM, an iterative algorithm based on the alternating direction method of multipliers (ADMM). The main subproblems admit closed-form or projection-based updates. Simulation results show that DANM-ADMM improves DOA estimation accuracy over benchmark methods while maintaining low computational complexity.
Keywords:
DOA estimation
reconfigurable intelligent surface
ADMM
decoupled atomic norm minimization (DANM)
non-line-of-sight (NLoS)
Journal
IF:
4.4
Papers:
1.2W
Citations:
2.2W
Organization
Cited Papers