Wavelet denoising for plane wave DOA estimation by MUSIC
R. Sathish, G. V. Anand · 2004
MUSIC is a widely used technique for plane wave direction-of-arrival (DOA) estimation, which is a problem of great interest in several applications. The performance of MUSIC degrades under low signal-to-noise ratio (SNR) conditions due to errors in estimating the data covariance matrix from finite data. The paper explores the possibility of employing the wavelet denoising technique to arrest the degradation in the finite data performance of MUSIC under low SNR. We propose the application of wavelet denoising to the noisy signal at each sensor to boost the SNR before performing DOA estimation by MUSIC. A comparative study of the finite data performance of MUSIC is presented for the undenoised and denoised data, and it is shown that denoising leads to a significant reduction in the bias and mean square errors (MSE) of the DOA estimates.