This paper presents a novel direction of arrival (DOA) estimation method via rotation spatial differencing technique that offers high resolution, robustness, and stable performance. To suppress external environmental noise and improve estimation accuracy, a new modified covariance matrix is constructed using a rotation matrix technique. Additionally, a spatial differencing matrix is built with neighboring subarrays to achieve signal decoherence. Due to the introduction of the new modified covariance matrix, the differencing matrix has a new feature without compromising the array flow pattern, leading to improved spatial differencing. Finally, a multiple signal classification (MUSIC) spectral search algorithm using the singular value decomposition (SVD) is applied to accurately localize both uncorrelated and coherent signals at once, which greatly facilitates the DOA estimation process. Experimental results demonstrate that the proposed method delivers superior DOA estimation performance, providing accurate and stable signal direction estimation.
1 October 2026 in IEEE Journal of Oceanic Engineering