Research On High-resolution Algorithm Of Direction Of Arrival Estimation

Spatial spectrum estimation is a significant research direction of array signal processing,and has been widely applied in a variety of fields ranging from radar, sonar, andcommunication etc. High-resolution Direction Of Arrival (DOA) estimation algorithms basedon subspace theory have attracted a great many experts investigating，because these methodshold excellent estimation performance. However, these high-resolution DOA algorithms holdhigh computational complexity, because of the covariance matrix computation and eigenvaluedecomposition, and these methods can be only work with independent signals and nointerference signals. In the electromagnetic environment, interference signals and coherentsignals are inevitable since the transmission path of electromagnetic wave is extremelycomplex. Hence, this paper focuses on the DOA estimation issue with interference signals andcoherence signals in multi-path environment, and the DOA estimation issue of theMultiple-Input Multiple-Output (MIMO) radar system.Firstly, the fundamental theory of spatial spectrum estimation is argued. Themathematical model is offered, and the signal model and array model of spatial spectrumestimation are introduced detailedly. Then the principle of basal spatial spectrum isinvestigated, and the estimation performances are summarized through simulations andcomparisons.Secondly, considering the interference signals in multi-path environment, a fastanti-interference DOA estimation algorithm is proposed. The forward recursion property ofMultistage Weiner FilterMSWF) and anti-interference principle of block matrix are investigated elaborately.Theory analysis and simulation results verify that the proposed method has low computationalcomplexity and the availability of anti-interference.In multi-path environment with coherent signals existing, after researching thede-coherent principles of spatial smooth algorithms, the merit and demerit of this class ofalgorithms are obtained by simulation results. A virtual array based on unitary transformationmethod for direction estimation of coherent signals is proposed. The proposed algorithmcombines the virtual array technique and unitary transformation. The mew method has no lossarray aperture and all of eigenvalue decomposition in the real space, so it has lowcomputational complexity. Simulation results verify that the proposed method has good de-coherent property, and better estimation performance than both spatial smooth algorithmsand virtual array algorithms.Finally, considering the DOA estimation for monostatic MIMO radar system, a fast DOAestimation algorithm is proposed. The signal subspace of monostatic MIMO radar system isobtained by the forward recursion of MSWF, and spatial spectrum function is constructed andsolved by transferred into root form. The proposed method avoids the eigenvaluedecomposition of covariance matrix and the spatial spectrum searching. So the computationalcomplexity is reduced tremendously. Simulation results demonstrate that the proposedalgorithm has good estimation performance.