Improvement of DOA Estimation Using Wavelet Denoising

The performance of direction of arrival (DOA) estimation such as the conventional beamforming (CBF) degrades progressively with the decreasing of signal to noise ratio (SNR). The performance of DOA estimation may be improved by employing a pre-processing method which enhances the SNR before performing DOA estimation. In this paper, wavelet denoising is applied in the estimation of the direction of arrival. The idea of this method is that prior to the estimation of the DOA, each of the sensor outputs is denoised according to wavelet denoising in order to enhance the input SNR, after that the denoised signals are used for the DOA estimation. Computer simulations illustrate that the proposed method generally gives a better DOA performance over the conventional methods.