A novel improved MUSIC algorithm by wavelet denoising in spatially correlated noises

In the presence of spatially correlated noises and/or at low SNR, the most popular multiple signal classification (MUSIC) method degrades greatly and even fails to estimate closely spaced signal directions-of-arrival (DOA's). We have proposed an improved DOA estimation method by wavelet denoising in the context of spatially correlated noises. The proposed approach denoises the received signals at each sensor in parallel and the estimates are obtained by applying MUSIC to the denoised data matrix. Simulation results, proving the output SNR enhancement and resolution improvement, are given.