Pix2Pix-based DOA Estimation with Low SNR
Xianglin Cao, Fuwei Wang, Bobo Yi, Zhoupeng Wei, Lu Liu · 2022
This paper proposes a direction of arrival (DOA) estimation method that can work at low signal-to-noise ratios (SNR). And the method is based on the classical model Pix2Pix in generative adversarial networks. Using the Pix2Pix graph-to-graph conversion capability, the covariance matrix with low SNR is converted to a covariance matrix with high SNR, enabling high-precision DOA estimation at low SNR. Experiments show that the root-mean-square error (RMSE) of multiple signal classification (MUSIC) is more than 30 times higher than that of the method at a SNR of -10dB.