Based On Uniform Circular Array For Doa Estimation Algorithms

Array signal processing, known with the two main parts - estimating direction of arrival (DOA) and digital beam forming (DBF), is an important branch of signal processing, and plays an important role in the military and civilian applications. In this dissertation, we discuss the direction-finding problem in the broadband signal model, based on the uniform circular array (UCA), which is famous with more absorbing direction-finding aspects compared with the uniform linear array (ULA). The main researchs are summarized as follows:1. First the classic super-resolution DOA estimator, such as MUSIC and ESPRIT, is discussed based on the ULA and UCA array data model for narrowband signals.2. The basic theories and algorithms of UCA are introduced, then extended to handle the direction-finding problem of the broadband signal. And the performance of the extended work is verified via computer simulation.3. The coherent signal subspace methods (CSM) of the broadband signal's DOA estimation are discussed in detail,and simulation results show that CSM can focus the whole signal band's energy to improve the signal to noise ratio.4. A new method is presented by jointly using the phase mode excitation of the uniform circular array (UCA) and CSM focusing technique. In the presented method, data is first transferred into phase-mode space where the ULA-like manifold structure can be obtained, then CSM focusing technique is used in the UCA phase-mode space to eliminate the manifold's dependence on frequencies of broadband sources, and DOAs of wideband sources can be estimated. The simulation results indicate the efficacy of the proposed method.