A DOA Estimation Algorithm for Coherent Signals Based on Vision Transformer
Aokun Kong, Yue Cui, Shuqin Wang, Yuezhou Jing · 2025
This paper presents a Direction of Arrival (DOA) estimation method for coherent signals using Vision Transformers. It combines Vision Transformers and CNNs, using real, imaginary, and phase components of the covariance matrix as multi-channel inputs. A self-attention mechanism extracts features, and a regression model is used as input to avoid the grid mismatch problem. Simulation results show the method outperforms existing approaches in accuracy, decorrelation, and runtime.