Research On Beamforming And Subarray Technology Of Large-scale Antenna Array

Beampattern synthesis has always been a core technology in the field of sensor array signal processing.This technology achieves the effect of controlling the distribution of received or transmitted electromagnetic wave power at different spatial angles by adjusting the feed excitation of each array element antenna in the array.With the development of science and technology,in practical engineering applications in recent years,the scale of the array is also developing with the trend of large-scale antenna array.However,the system complexity and the cost of the feeding network are gradually rising with the increasing of the scale of the antenna array,which makes the application research of large-scale antenna arrays challenging.This thesis focuses on the application research of large-scale antenna arrays,and aims to reducing the design cost of large-scale antenna arrays and simplifying the computational complexity and design complexity of the system.Through the research on sparse beamforming technology and subarray classification technology,this thesis is expected to improve system performance and save system cost of the large-scale antenna arrays.The main research of this thesis is as follows:1.Based on the sparse beamforming method of convex optimization,this thesis proposes a sparse array beamforming scheme with sidelobe control.In particular,the sparsity of the antenna array is quantified by using the ₁l norm of the designed weight vector.Moreover,the tradeoff between the sparsity of array and sidelobe level is achieved by changing the lower and upper magnitude response bound constraints in the sidelobe region.Then,a nonconvex optimization problem is formulated with the consideration of the magnitude response bound deviation constraints in the mainlobe region.By introducing auxiliary variables,the alternating direction method of multipliers(ADMM)framework is introduced to decompose the original problem into subproblems,and the weight vector and auxiliary variables are obtained iteratively.Finally,theoretical simulations show that the presented algorithm can realize the sparse array design and sidelobe level control.2.This thesis summarizes the subarray classification method of the array,and mainly analyzes the side-by-side type or the localized architecture and the interleaved structure of the uniform subarray.Moreover,this thesis solve the mathematical modeling of the localized architecture and the interleaved structure respectively.By establishing a mathematical model,the problem of subarray classification is transformed into a corresponding convex optimization problem to solve.Finally,through numerical theoretical simulations,the performance of the localized architecture and the interleaved structure are analyzed,as well as their application scenarios are analyzed too.3.Based on the existing interleaved structure classification method,this thesis proposes a method which separates the digital channels from the analog channels of subarrays.The analog parts design the analog weight coefficients with the same amplitude,and the digital parts designs the digital weight coefficients in the real number domain.The advantage of the subarray structure is that in the actual hardware implementation,only a phase shifter needs to be introduced at the analog parts without using an amplifier,and at the digital parts,only the amplitude of the signal needs to be controlled without adjusting the phase,so that this method greatly saves the hardware cost which the whole system is required.