Research On Antenna Array Method And Implementation By The GPU

The disposal structure of antenna array has a very important role in many engineering practice. Through to the antenna array structure planning and the arrangement of antenna array element position, we can make the antenna gain the ideal shape beam and the low side lobe level and strong directivity, and so on. The maximum relative side lobe level of the antenna array is also a performance that is often investigated. How to allocate the array elements of the antenna array quickly and efficiently, so that the array has the maximum relative side lobe level as low as possible,is the core of the study.The main work and research contents of this paper are as follows:1. The optimal mathematical model of sparse linear array and sparse rectangular plane array is established, and the solution of the optimal position distribution of array element is discussed. And the algorithm is used to simulate the sparse linear array by the enumeration algorithm preliminary.2. The genetic algorithm (GA) method of strategy optimization of sparse antenna array is discussed. The Simulation results of parse linear array and sparse rectangular plane array is obtained, and the optimization effect under different conditions is analyzed.3. The structure of the molecular sparse arrays is studied. First, the optimization model of the structure is established. And the standard genetic algorithm is modified according to this model, which solves the problem of lack of flexibility and slow speed of convergence of the arrangement of sparse arrays. Experiment result shows that in the case of sparse linear array, the array of molecular structures is optimized to obtain a lower maximum relative side lobe level under this study's experimental parameters.4. In this paper, a GPU implementation method based on CUDA programming is studied in the realization of genetic algorithm optimization antenna,which is based on the computational complexity and parallelism of genetic algorithm. The CUDA programming model is given, and the parallel genetic algorithm for antenna array is designed and implemented. At the same time, the program is optimized for GPU and CUDA structure model to achieve high utilization rate. Finally, through the simulation to get the CPU and GPU platform optimization time, verify the High-efficiency of GPU implementation.