Research On Beamforming Technology Based On Dynamic Metasurface Antenna Arra

With the rapid development of large-scale multiple input multiple output(MIMO)technology in the field of wireless communication and radar,the requirements for the cost,size and power consumption of antenna array are higher in practical application.As a compromise between low-cost analog array and high-performance digital array,analog-digital hybrid array significantly reduces the hardware cost and system implementation complexity of antenna array at the cost of small beamforming(BF)performance loss.It has become one of the core technologies of large-scale MIMO.Dynamic metasurface antenna(DMA)is a new type of antenna derived from metamaterials.It has many advantages,such as small size,low cost,low power consumption,high gain and so on.Since it was proposed,it has attracted a high degree of research interest of scholars at home and abroad.DMA array is a new analog-digital hybrid array structure based on DMA.Its analog beamforming part is realized by DMA.The transmission of electromagnetic wave(EM)is controlled by using the physical characteristics and arrangement of metamaterial array elements,which can effectively realize the compromise between the beamforming performance of antenna array and the complexity of hardware implementation.Array pattern design and synthesis is a basic research content of array transmit beamforming technology.In practical application,such as radar,communication,automatic driving and other different application scenarios,the design of array pattern has an important impact on the performance of array system.In view of this,this thesis mainly studies the transmit array pattern design of DMA array.Based on different array pattern design criteria,this thesis respectively proposes two kinds of hybrid transmit beamforming design method for DMA array,and their core is the joint optimization design of analog weights and digital weights.Due to the coupling between digital weight and analog weight in DMA array,and the analog weight meets a nonlinear amplitude phase coupling constraint,hybrid transmit beamforming is essentially a nonconvex optimization problem,which is difficult to be solved directly by existing methods.Based on the idea of alternating optimization,this thesis decomposes the analog-digital weight joint optimization problem into several subproblems,and then iteratively solves them.For DMA array hybrid beamforming,the pattern design contents and innovations are as follows:1.The hybrid transmit beamforming design of DMA array under the expected pattern fitting criterion is studied.Under the artificially desired pattern,the analog weight and digital weight of DMA array are designed to minimize the weighted sum of the square error between the amplitude of the pattern to be designed and the amplitude of the expected pattern.Among them,the introduction of pre weighting factor can control the depth of the null region and fit the null region more accurately,Finally,the hybrid transmit beamforming of DMA array in different application scenarios is realized.2.Aiming at the disadvantage that the sidelobe level needs to be manually specified in the array pattern design under the expected pattern fitting criterion,the hybrid beamforming design of DMA array based on the minimum peak sidelobe ratio maximization criterion is studied.This criterion does not specify the desired pattern under the amplitude and phase constraints of DMA array simulation weights,takes the minimum main lobe level and peak side lobe level as reference points,and directly designs the DMA array hybrid weights to maximize the ratio of minimum main lobe level to peak side lobe level,so as to consider the joint control of main lobe and side lobe and realize the normalized minimum side lobe level characteristics,By suppressing the peak sidelobe level and increasing the main lobe level,the DMA array can achieve a sufficiently low peak sidelobe level beam pattern under the condition of meeting the basic waveform requirements.