| Array signal processing technology is an important subject in the field of signal processing, which is widely used in communication, radar, sonar and electronic countermeasure etc. Super-resolution DOA (Direction of Arrival) estimation algorithm is a hot topic in research of array signal processing, which is mainly to solve the problem of high-precision and super-resolution direction estimation of the spatial signal. Therefore, we investigate super-resolution DOA estimation algorithm in this paper, in combination with specific research project.First of all, for one-dimensional DOA estimation of the uniform linear array, a novel DOA estimation method was proposed based on generalized Cauchy distribution and MAP. The algorithm is different from the traditional super resolution DOA estimation algorithm, which is based on prior distribution. Compared with MUSIC algorithm, the new algorithm has the advantages of high precision and sparse space spectrum without using spectrum scan.Secondly, in order to meet the new demands of 2-D DOA estimation including azimuth angle and elevation angle,2-D DOA estimation based on the uniform circular array was researched.2-D MUSIC algorithm of UCA is proposed on basis of one-dimensional MUSIC. In addition, using weighted spectral technique, C-UCA-ESPRIT algorithm is developed in phase mode excitation space. This new method has low computational complexity, and automatically matches the DOA parameters keeping lower bias.Finally, an improved method of DOA estimation based on QR decomposition is proposed in order to meet the demands of real-time system of DOA estimation. The new algorithm uses Householder transformation and QR decomposition to solve eigen-decomposition problem of subspace DOA estimation algorithm, and is more suitable for implementation of the hardware platform. Eventually, the presented algorithm is implemented on DSP hardware platform, and meets the requirements of super resolution and real-time system. |
PDF Full Text Request