Research On Antenna Array Element Position Optimization And Beamforming Technology

The antenna array can effectively control array response and has a wide application in radar,communications and other fields.Array element position optimization and pattern synthesis are based on optimizing the number and spacing of array elements and designing antenna array weight vector to optimize the antenna array pattern,which are of great research significance.This thesis studies the array element position optimization and beamforming techniques with rectangular planar arrays and hemispherical conformal arrays as the background.The main research contents are as follows:Firstly,the traditional planar array receiving beamforming and conformal array transmitting beamforming are modeled,and the rectangular planar array and hemispherical conformal array models are established.Secondly,the array elements positions of the rectangular planar array are optimized based on the particle swarm optimization algorithm(PSO).This thesis studies the thinned array and the sparse array separately.The Boolean PSO has been improved so that the sparsity rate of the array can be controlled to achieve a rectangular thinned array with the constraint of the number of array elements.Given the problems in the existing rectangular planar sparse array algorithms,an optimization method for rectangular sparse array based on the self-adjusting mapping rule is proposed.The algorithm can effectively reduce the peak sidelobe level under multiple constraints,and its performance is better than the thinned array and the existing sparse array algorithms under the same conditions.Then,the conformal array transmitting beam array optimization is studied.Based on the traversal law of the main beam gain,the deployable array elements are obtained,which effectively solves the problem of the offset of the main beam of the transmission pattern.On the basis of deployable array elements,the multi-objective PSO is applied to the transmitting single-beam array optimization,and the simulation shows that its performance is better than that of the binary PSO.To give full play to the advantages of all-digital arrays and improve the utilization rate of array elements,the thesis proposes a double PSO.The algorithm obtains the optimal array for each beam by the multi-objective PSO and determines the attribution of multiplexing array elements through the multivariate PSO to realize the simultaneous transmitting of multi-beam array optimization based on conformal arrays,avoiding the multiplexing of array elements.Finally,three pattern synthesis algorithms to further improve the performance of the conformal array transmitting pattern are proposed in this thesis.Based on convex optimization theory,a pattern synthesis algorithm based on the maximum output signalto-noise ratio is proposed,and this algorithm realizes pattern synthesis by solving the convex optimization problem to obtain an optimal weight vector.Except that,in order to balance the performance and computational efficiency,this thesis uses adaptive theory to propose pattern synthesis via pre-weighted orthogonal projection decomposition and a pattern synthesis algorithm based on linearly constrained minimum variance(LCMV)complementary decomposition,respectively.By decomposing and adjusting the optimal weights,the weighted pattern approximates the desired pattern.This thesis finds that the performance of the three pattern synthesis algorithms is similar from the simulation results,and they can achieve pattern synthesis of conformal array to meet specific performance requirements.The latter two algorithms avoid solving convex optimization problems,having higher computational efficiency and more application value.