Study On Analysis Of Array Errors And Calibration Techniques

Array signal processing, considered as an important method for the spatial information processing, develops rapidly. It usually applies to electronic surveillance, radar, communications, sonar, seismology, radio astronomy and other fields. Most of the existing array signals processing methods need to know about the accurate array manifold. The performance of methods will be heavily influenced by array errors, such as the amplitude, phase errors and array element position error, etc. Therefore, the research on the methods of array calibration is a longer but newer issue.In this thesis, the effect of errors and two methods on array calibration are studied. The following major content is:First of all, the mathematical model and high-resolution direction finding algorithm about uniform linear array are presented. It is analyzed that how to evaluate the influence of amplitude/phase errors and array element position error. In addition, the performance of DOA estimation, both on the far field plane wave model and on the near field spherical sound wave model, are researched.Secondly, calibration with signal source based on method of covariance Eigen-Decomposition, can calculate accurate array manifold. But due to the large computational algorithm, an alternative method based on the Propagator Method algorithm is proposed. This method can effectively solve that problem and apply to real-time computing. Finally, algorithm for compensating near-field spherical wave to far-field plane wave is studied.At last, the experiment using four electret microphones array show that two methods of array calibration are viable.In this thesis, the effect of errors and the methods on array calibration are research. The results show Propagator Method algorithm reduces the computational complexity and helps improve the real-time.