Robust Beamforming And DOA Estimation Of Wideband Signals

Facing with the rich and complex marine environment,the ability to develop marine resources,marine safety precautions,and the use of marine equipment has become an important indicator of the country's overall national strength.With the rapid development of science and technology and the rapid progress of modern information technology,array technology is widely used in marine equipment.Therefore,the research on array related theory has reached a new height.Beamforming and Direction of Arrival(DOA)as a key technologies involved in array signal processing,are always a hot issue in signal processing.Beamforming plays an important role in various fields such as wireless communications,military applications,and the relevant technologies DOA estimates have been widely used in military,marine and other major fields in recent years.The wideband signal carries the target echo and carries a large amount of useful information.It also puts forward more stringent requirements on the accuracy and real-time performance of the DOA estimation of the wideband signal.Compared with the uniform array(ULA),the number of sources and the number of array elements in the non-uniform array have no clear requirements in the non-ULA,the degree of freedom of the array can be improved.In recent years,studies based on non-ULA have received extensive attention?The research in this dissertation mainly focuses on the following aspects:improving the robustness of beamforming and DOA estimation based on non-ULA with wideband signals.At the same time,it needs to consider the cost of related equipment in real-world application scenarios and the real-time performance of the technology.The main works of this paper are as follows:(1)Since the insufficient for traditional beamforming methods,this study mainly discusses how to improve the performance of beamforming from the standpoint of robustness.A robust beamforming method based on iterative estimation is proposed.Improve the accuracy of the estimation by further improving the estimated steering vector and improve the robustness of beamforming performance.After the correlation algorithm is improved,Newton method is introduced to perform secondary estimation on the steering vector to improve the accuracy of the estimation and reduce the computational complexity,thus improving the robustness of beamforming.(2)As to the problems with DOA estimation in the actual application environment,it is more common to have more than one target simultaneously.The key issue of the DOA estimation is the ability to accurately determine the number of unknown sources.By introducing a non-uniform array and comparing with a uniform linear array,the arrival direction of multiple sources can be determined by using fewer array elements.The sparse representation method in compressed sensing is used to improve the accuracy of the estimation.And the conjugate gradient method is also used to realize fast estimation of wideband signals and achieve the purpose of real-time implementation.(3)This paper presents combining DOA estimation with wideband signals and DOA estimation,for the difference between ULA and non-ULA.Covariance matrix is reconstructed by specific methods in compressed sensing to avoid Considering Array Continuity,transform the DOA estimation problem into the solution to the optimal solution.Based on the actual situation,compare the coprime array with other types of non-ULA.The simulation results show that the correctness of the method,and improving the freedom of the array by using coprime array,increase the aperture of the array and reduce the computational complexity.(4)In combination with the characteristics of the array,relevant experiments were conducted in the Wuyuan Bay area of Xiamen,and the characteristics of underwater acoustic channels in the relevant sea areas were analyzed.Different methods were compared to further verify various methods.