空间谱估计算法研究-应用数学专业毕业论文.docxVIP

华中科技 大学硕士 学 华 中 科 技 大 学 硕 士 学 位 论 文 II II Abstract In passive location system, the DOA estimation is an important issue which will decide whether the object could be located or not. As an important area in array signal processing, DOA estimation draws more and more attention because of its high performance and wide usage. This paper focuses on the two problems of how to reduce the computation load of eigen-subspace algorithm and the DOA algorithm for coherent signals. The main contents of this research are as follows. The research on the real-time realization of eigen-subspace algorithm. Based on the basic MUSIC method and its characteristic, we investigate Beam-space MUSIC method and Root-MUSIC method; give the introduction of the Unitary operation on array steer vector and data covariance matrix and the introduction of DFT method which introduces the DFT transform into the DOA estimation; propose a new fast subspace algorithm for DOA estimation. The Matlab simulation results show that the Beam-space method, the Root method, Unitary operation and fast subspace DOA method all can reduce the computation load of eigen-subspace algorithm effectively. But the performance of Unitary operation is not very good. The DFT method has a low computation complexity and a little operation time because there is only a snapshot data involved. And the performance is not very good due to the limited data. The research on the DOA algorithm for coherent signals. We introduce the characteristic and decorrelation performance of spatial smoothing(SS) method; introduce the Blind estimation method which makes a frequency spectrum estimation on the generalized steer vector of coherent signals and gets the DOA information indirectly; propose a new method—DSM method for DOA estimation. Essentially, DSM method only bases on the characteristic of sensor array and doesn’t rely on the eigen-structure of sample data subspace. And it can be widely used in different areas becau