Research On Low-complexity Three-dimensional Source Location Algorithms Based On Sparse Arrays

Source location technology is one of the most important applications in the field of array signal processings,whose core function is to estimate the location parameters of the source.The three-dimensional(3D)source location,with the ability to measure the elevation,azimuth and range parameters of the source at the same time,is important and widely med in practice in the source location fields,such as 5G millimeter-wave communications,wireless sensor networks,sound source location and seismic exploration.In addition,the high-precision 3D source location could accurately help the MPT system to perform energy directional transmission,which has great potential for improving the transmission efficiency of the MPT.However,problems,such as complex array,cumbersome guiding vector and high complexity,restrict the development and use of 3D source location algorithms.Hence,in the dissertation,the low-complexity 3D source location algorithm is investigated from two aspects,the array model and source location algorithm,respectively the main work is as follows:1)A Sparse Planar Array based on the Coprime Linear Array Coprime Shift(CLACSSPA)and a Coprime Cross Array(CCA)are constructed.Both the CLACS-SPA and CCA are two-dimensional arrays based on coprime,whose complexity of the array structure is all much lower than that of a uniform planar array with the same aperture.In particular,the CCA array achieves extremely lower array structure complexity,in comparison with CLACS-SPA.At the same time,both CLACS-SPA and CCA arrays can complete highprecision and long-range 3D source location tasks without ambiguity,where the CLACSSPA array has better location performance than CCA.2)A Spherical Conformal Cross Array(SCCA)is constructed.The elements on the array are distributed with isometric angle on two arc-shaped linear arrays,and these two arc-shaped linear arrays are respectively on the intersections of two mutually perpendicular planes passing through the center of the sphere and the ball.The SCCA can not only well attach to the surface of the sphere to improve the space utilization of the array,but also has the structure characteristic of sparse array with low complexity.The steering vector matrix structure of the array is deduced,and show a 3D source location model based on the array is established.Simulation analyses show that the constructed array has higher source location accuracy compared with the CCA array with the same aperture and the same number of antennas.In addition,the impacts of the parameters on the location performance of the array,including radius,radian and element spacing in the array are obtained.3)Two low-complexity 3D source location algorithms are studied.The 3D Rank Reduced Multiple Signal Classification(RARE-MUSIC)and the 3D Rank Reduced Propagator Method(RARE-PM)are based on the core idea of rank reduced techniques,and are decomposed by using Taylor series method,to separate the range item from the source steering vector.In this way,the simultaneous spectral peak searches for the elevation,azimuth and range,are replaced by one two-dimensional spectral peak searching for the elevation and azimuth and one one-dimensional spectral peak searching for the range term.The simulation results show that both the computational complexity and the errors in the errors in the estimation of the 3D parameters of the source are reduced greatly in algorithms.