Research On High-Resolution Direction-of-Arrival Algorithm Of 3D Moving Platform Mounted Array

The direction-of-arrival(DOA)algorithm is an important part of the array signal processing,which can obtain the target information of the received sigal.With the enrichment of people’s understanding of the ocean,the status of the ocean in resource exploitation,scientific research,military affairs and other aspects has become important,and more and more Underwater Unmanned Vehicles and Autonomous Underwater Vehicles have begun to be applied in underwater scientific research and reconnaissance.However,the array mounted on a small motion platform can only select a small linear array or a conformal array,which puts forward the high-resolution requirements for DOA.In addition,the movement of the small motion platform is complex,so it is necessary to perform 3D azimuth correction technology.The paper takes the small motion platform equipped with array received signal as the research object,researches the high-resolution DOA algorithm,and improves the algorithm for 3D azimuth correction according to the actual physical background.Firstly,the array signal receiving model is established,the Richardson-Lucy(R-L)deconvolution DOA algorithm based on uniform linear array is studied,and a frequencynormalized R-L deconvolution DOA algorithm is proposed.In broadband processing,the R-L algorithm needs to calculate the point spread function dictionary for each frequency point respectively,which leads to the increase of the computation amount.Therefore,a frequencynormalized R-L deconvolution algorithm is proposed,which can effectively save computing and storage resources without affecting the performance of the algorithm,and can use the periodicity of the frequency-normalized algorithm to carry out simple and convenient,which is verified by simulation analysis and sea test data.Secondly,the paper studies the Sparse Bayesian Learning(SBL)DOA algorithm and proposes an SBL(Ex＿SBL)DOA algorithm based on fixed noise estimation.The performance of the SBL DOA algorithm is directly related to the hyperparameter noise power estimation.This estimation of the existing SBL algorithm must be obtained through iteration,which requires a large amount of calculation.Therefore,the paper uses the physical significance of conventional DOA to give an Ex＿SBL DOA algorithm,which can improve the processing effect and save computing resources when the broadband is low in signal-to-noise ratio,and is verified by simulation analysis and sea test data.Finally,the high-resolution DOA algorithm is improved and applied to the irregular smallscale array carried by the three-dimensional motion platform.Firstly,the principle of the 3D correction algorithm is introduced,and a DOA algorithm for 3D azimuth correction is proposed.SBL DOA technology can be directly applied to irregular small arrays,but R-L deconvolution DOA technology can only be applied to shift-invariant arrays,so an R-L deconvolution DOA technology suitable for shift-variable arrays is introduced.Finally,the high-resolution DOA algorithm for 3D azimuth correction is simulated and verified with experimental data.In summary,this paper studies the high-resolution DOA algorithm based on the threedimensional motion platform mounted array.The simulation and test results show that the small motion platform mounted small array can obtain the true earth bearing information of the target and better resolution effect when using the 3D azimuth modified high resolution DOA algorithm.