Enhanced DOA Estimation Method Based On Known Waveforms

Direction-of-arrival(DOA)estimation is an important research content in the array singal processing,and has been applied to a variety of civil and military fields,such as communications,nevigation,radar,sonar,etc.In the past decades,a great many DOA estimation algorithms have been proposed,and these techniques generally consider the source signals to be random process or unknown deterministic process,which utilize only spatial properties such as the auto-correlation matrix,without taking into account any prior information.In practice,there are abundant available prior information on signals of interest,such as signal waveform,characteristics of cyclostationary,non-circular and constant modulus in wireless communication,active radar or sonar.Previous papers have proven that they can be exploited to significantly improve DOA estimation accuracy,simplify computational complexity and eliminate interfering signals.Therefore,it is of great significance to study how to use these prior information to improve DOA estimation algorithms.This paper mainly studies the DOA estimation methods using known waveforms,focusing on incoherent and coherent signals,and analyzes the corresponding Cramér-Rao bound(CRB)characteristics.Considering the actual demands,the method configuration is extended from linear array to tridimensional(3D)array.The main work and contributions of this thesis include the following three aspects:1.For incoherent signals,it proposed an interferometer-based DOA estimation algorithm with known waveforms.The algorithm used the estimation of baseline phase difference to obtain a closed form DOA solution.Besides,it derived the CRB of uniform linear array and uniform circular array and their characteristics are analyzed.The proposed algorithm is suitable for arbitrary linear array and 3D array,and the calculation is very simple without complex operations such as spectral peak search or eigenvalue decomposition,and the performance can reach the CRB under most conditions.2.It proposed a coherent signals DOA estimation algorithm with known waveforms using sparse Bayesian learning.Since the spatial sparsity of DOA,the algorithm transformed a DOA estimation problem into solving the sparse coefficient vector of spatial feature vector.The proposed algorithm is suitable for arbitrary linear array and 3D array and avoids the performance degradation of traditional de-coherence algorithm,which is caused by array aperture loss of spatial smoothing,and the performance is almost the same as Maximum likelihood estimator,which both reach the CRB.3.It summarized the CRBs of linear array and 3D array with unknown or known signal waveform.Specifically,it derived the CRB of two-dimensional DOA estimation of a 3D array with known waveforms.And it analyzed the characteristics of CRB with known waveforms when signals are uncorrelated.In the case of only one incident signal,the CRBs of unknown and known waveforms are compared,and the performance improvement by waveform information is quantitatively explained.