Research On Key Technologies For Spherical Near-Field Measurement Of Vehicular Antenna And 5G Base Station

The flourishing development of communication technology facilitates the rapid growth of the antenna industry,leading to an intensifying demand for fast and accurate measurement of radiation parameters,so as to deliver more reliable reference for antenna design.Meanwhile,the occurrence of high performance antennas also puts forward enhanced requests for measurement technology.For instance,as for the vehicular antenna and 5G base station,since the measurement of the individual antenna fails to precisely reflect its radiation performance under normal operation state,the automobile and the active antenna should be regarded as a whole,rendering the electrical size and the far-field distance of the antenna under test extremely large.Compared with the far-field and CATR methods,spherical near-field measurement has significant superiority in terms of dimension and far-field precision,therefore,this paper concentrates on the spherical near-field measurement technique.The specific work and innovation are as follows:1.To address the issues of severe far-field errors and extremely high oversampling rates induced by the strict equiangular sampling combined with the scattering matrix method,this paper is devoted to ameliorating the pervasive spherical near-field measurement technique by means of both the sampling grids as well as the transformation algorithm.First,four non-equiangular grids are imported to alleviate the serious sampling redundancy under the equiangular form.Second,in the aspect of the transformation algorithm,a new approach is proposed to solve the transmission coefficients utilizing the inverse matrix method.Numerical results demonstrate its potent flexibility for non-equiangular grids and validity in promoting the far-field accuracy and minimizing the sampling rate.Simultaneously,the lower limit of the oversampling rate is approximated.Third,for continuing decreasing the quantity of the samples required to restore the far field of the geometrically symmetric antenna,on the basis of the priori knowledge that the corresponding transmission coefficients exhibit sparsity,in conjunction with the compressed sensing,the precise reconstruction of the far field under under-sampling conditions can be achieved in this paper.2.A spherical near-field measurement technique adapted to the vehicular antenna is presented.Due to the restriction of the automobile,the vehicular antenna measurement will be quite different from the ordinary one in two aspects: the setup of the scan center and the test arrangement.First,targeting the problem that the spherical modes and the samples requested to retrieve the corresponding transmission coefficients are dramatically enlarged due to the deviation of the vehicular antenna from the scan center,a compensated algorithm is developed that can substantially diminish the oversampling rate requisite for the offset-mounted antenna.Second,with respect to the site configuration,a flat,mechanically stable metallic floor is incorporated to narrow the test space and facilitate vehicle movement.Regarding the above test scenario,the suggested algorithm is capable of handling both the impact of the metallic ground on the radiated field and the offset compensation.At a lower oversampling rate,the acquisition of the precise three-dimensional radiation pattern of the offset-mounted antenna in free space can be accomplished merely in accordance with the complex superimposed field on the hemisphere above the ground.3.A phase-less spherical near-field measurement technique applicable to 5G base station is suggested.To tackle the difficulty of inaccessible stable reference phase in OTA(Over the air)test,this paper models the computation of the transmission coefficients as a phase retrieval problem and cooperates with a near-field bi-spherical sampling strategy to compensate for the inherent deficit of phase information.Incorporating the far-field precision and other indicators,an optimal phase retrieval algorithm is identified.In addition,an adapted sampling method is outlined,concerning about the radius difference,the grids and the polarization.In contrast to the bi-spherical iterative method,numerical results substantiate the superior performance of the proposed scheme in terms of the sampling rate,the far-field accuracy and the operational efficiency.