Shaped Beam Synthesis Of Satellite-borne Planar Array

To solve the problem of LEO-satellite-borne multi-beam planar arraybeamforming, genetic algorithrn (GA) was used for beam-cell arrangement andshaped beam pattern synthesis.A genetic algorithm was developed to optimize the number and size of cells for acircularly symmetric grid. A phased array antenna with circular aperture and cosθelement power pattern was used. The genetic optimizer was found to quickly produceoptimal cell layouts for arbitrary altitude, field of view, and directivity requirements.It was also a very good way to quickly and accurately determine the number of beamsneeded for a particular set of requirements.A planar array antenna with arbitrary geometry synthesis technique based ongenetic algorithm is discussed. This approach avoids coding/decoding and directlyworks with complex numbers to simplify computing program and to speed upcomputation. This approach uses two crossover operators that can overcomepremature convergence and the dependence of convergence on initial population.Simulation results show that this method is capable of synthesizing complex patternshapes of planar arrays with arbitrary geometry and can realize good sidelobesuppression at the same time.As regards to algorithm, the paper presented a new method, which combines GAand BFGS; genetic algorithm based on niches is also used. These methods found anew road in reducing searching space and computation load and in speeding upconvergence. According to the features of complex coding GA, we modified thestandard GA. This modified GA leads to a low dependence on initial populations andan enhancement of global convergence performance compared with the existingcomplex coding GA.